twitter.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uv7qHzIaVUwf"
      },
      "source": [
        "### Objective\n",
        "Twitter wants to automatically tag and analyze tweets for better understanding of the trends and topics without being dependent on the hashtags that the users use. They want to create a system of recognizing important content of the tweets using Named Entity Recognition (NER) to locate and recognise named entities."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "oeeu6I_p-rZx"
      },
      "outputs": [],
      "source": [
        "# Import Libraries\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "import tensorflow as tf\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OODxyW9oVaKq"
      },
      "source": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3ISiBySsVZCT"
      },
      "source": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8x4l3cbzVY1g"
      },
      "source": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AQQB_30IVYjN"
      },
      "source": []
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oQz8u3NgAUj7",
        "outputId": "03577114-75ae-4346-b2bf-988edd61a24e"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Collecting datasets\n",
            "  Downloading datasets-2.14.0-py3-none-any.whl (492 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m492.2/492.2 kB\u001b[0m \u001b[31m7.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting transformers\n",
            "  Downloading transformers-4.31.0-py3-none-any.whl (7.4 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.4/7.4 MB\u001b[0m \u001b[31m36.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.10/dist-packages (from datasets) (1.22.4)\n",
            "Requirement already satisfied: pyarrow>=8.0.0 in /usr/local/lib/python3.10/dist-packages (from datasets) (9.0.0)\n",
            "Collecting dill<0.3.8,>=0.3.0 (from datasets)\n",
            "  Downloading dill-0.3.7-py3-none-any.whl (115 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m115.3/115.3 kB\u001b[0m \u001b[31m15.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (from datasets) (1.5.3)\n",
            "Requirement already satisfied: requests>=2.19.0 in /usr/local/lib/python3.10/dist-packages (from datasets) (2.27.1)\n",
            "Requirement already satisfied: tqdm>=4.62.1 in /usr/local/lib/python3.10/dist-packages (from datasets) (4.65.0)\n",
            "Collecting xxhash (from datasets)\n",
            "  Downloading xxhash-3.2.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (212 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m212.5/212.5 kB\u001b[0m \u001b[31m21.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting multiprocess (from datasets)\n",
            "  Downloading multiprocess-0.70.15-py310-none-any.whl (134 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m134.8/134.8 kB\u001b[0m \u001b[31m12.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: fsspec[http]>=2021.11.1 in /usr/local/lib/python3.10/dist-packages (from datasets) (2023.6.0)\n",
            "Requirement already satisfied: aiohttp in /usr/local/lib/python3.10/dist-packages (from datasets) (3.8.4)\n",
            "Collecting huggingface-hub<1.0.0,>=0.14.0 (from datasets)\n",
            "  Downloading huggingface_hub-0.16.4-py3-none-any.whl (268 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m268.8/268.8 kB\u001b[0m \u001b[31m25.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from datasets) (23.1)\n",
            "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.10/dist-packages (from datasets) (6.0.1)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from transformers) (3.12.2)\n",
            "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.10/dist-packages (from transformers) (2022.10.31)\n",
            "Collecting tokenizers!=0.11.3,<0.14,>=0.11.1 (from transformers)\n",
            "  Downloading tokenizers-0.13.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.8 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.8/7.8 MB\u001b[0m \u001b[31m90.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting safetensors>=0.3.1 (from transformers)\n",
            "  Downloading safetensors-0.3.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.3/1.3 MB\u001b[0m \u001b[31m70.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: attrs>=17.3.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (23.1.0)\n",
            "Requirement already satisfied: charset-normalizer<4.0,>=2.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (2.0.12)\n",
            "Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (6.0.4)\n",
            "Requirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (4.0.2)\n",
            "Requirement already satisfied: yarl<2.0,>=1.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (1.9.2)\n",
            "Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (1.4.0)\n",
            "Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (1.3.1)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.10/dist-packages (from huggingface-hub<1.0.0,>=0.14.0->datasets) (4.7.1)\n",
            "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets) (1.26.16)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets) (2023.5.7)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets) (3.4)\n",
            "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas->datasets) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas->datasets) (2022.7.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas->datasets) (1.16.0)\n",
            "Installing collected packages: tokenizers, safetensors, xxhash, dill, multiprocess, huggingface-hub, transformers, datasets\n",
            "Successfully installed datasets-2.14.0 dill-0.3.7 huggingface-hub-0.16.4 multiprocess-0.70.15 safetensors-0.3.1 tokenizers-0.13.3 transformers-4.31.0 xxhash-3.2.0\n",
            "Collecting tensorflow-addons\n",
            "  Downloading tensorflow_addons-0.21.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (612 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m612.1/612.1 kB\u001b[0m \u001b[31m7.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from tensorflow-addons) (23.1)\n",
            "Collecting typeguard<3.0.0,>=2.7 (from tensorflow-addons)\n",
            "  Downloading typeguard-2.13.3-py3-none-any.whl (17 kB)\n",
            "Installing collected packages: typeguard, tensorflow-addons\n",
            "Successfully installed tensorflow-addons-0.21.0 typeguard-2.13.3\n"
          ]
        }
      ],
      "source": [
        "# Install Libraries\n",
        "!pip install datasets transformers\n",
        "!pip install tensorflow-addons"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "hOB7b1JlFMiI"
      },
      "outputs": [],
      "source": [
        "# Load data\n",
        "def load_data(filename: str):\n",
        "  with open(filename, 'r') as file:\n",
        "    lines = [line[:-1].split() for line in file]\n",
        "  samples, start = [], 0\n",
        "  for end, parts in enumerate(lines):\n",
        "    if not parts:\n",
        "      sample = [(token, tag)\n",
        "        for token, tag in lines[start:end]]\n",
        "      samples.append(sample)\n",
        "      start = end + 1\n",
        "  if start < end:\n",
        "    samples.append(lines[start:end])\n",
        "\n",
        "  return samples\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "XPxcdKRtr1Lh"
      },
      "outputs": [],
      "source": [
        "train_file = '/content/wnut 16.txt.conll'\n",
        "test_file = '/content/wnut 16test.txt.conll'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "ZSYodvKPFbD9"
      },
      "outputs": [],
      "source": [
        "train_samples = load_data('/content/wnut 16.txt.conll')\n",
        "test_samples = load_data('/content/wnut 16test.txt.conll')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "p-DY6PBqUr-5"
      },
      "source": [
        "#### The CoNLL format is a text file with one word per line with sentences separated by an empty line. The first word in a line should be the word and the last word should be the label ."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "zoaXEZMCGT5i"
      },
      "outputs": [],
      "source": [
        "samples = train_samples + test_samples\n",
        "schema = ['_'] + sorted([tag for sentence in samples\n",
        "                            for _, tag in sentence ])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vyBO52SQGdkd",
        "outputId": "20c85369-9546-470c-9805-e8658e4b46ea"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[('Made', 'O'),\n",
              " ('it', 'O'),\n",
              " ('back', 'O'),\n",
              " ('home', 'O'),\n",
              " ('to', 'O'),\n",
              " ('GA', 'B-geo-loc'),\n",
              " ('.', 'O'),\n",
              " ('It', 'O'),\n",
              " ('sucks', 'O'),\n",
              " ('not', 'O'),\n",
              " ('to', 'O'),\n",
              " ('be', 'O'),\n",
              " ('at', 'O'),\n",
              " ('Disney', 'B-facility'),\n",
              " ('world', 'I-facility'),\n",
              " (',', 'O'),\n",
              " ('but', 'O'),\n",
              " ('its', 'O'),\n",
              " ('good', 'O'),\n",
              " ('to', 'O'),\n",
              " ('be', 'O'),\n",
              " ('home', 'O'),\n",
              " ('.', 'O'),\n",
              " ('Time', 'O'),\n",
              " ('to', 'O'),\n",
              " ('start', 'O'),\n",
              " ('planning', 'O'),\n",
              " ('the', 'O'),\n",
              " ('next', 'O'),\n",
              " ('Disney', 'B-facility'),\n",
              " ('World', 'I-facility'),\n",
              " ('trip', 'O'),\n",
              " ('.', 'O')]"
            ]
          },
          "execution_count": 13,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "train_samples[1]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IDG6hzzqW5XU",
        "outputId": "96d3b41e-820f-48e7-ce0b-a9b4e5026a52"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "108376"
            ]
          },
          "execution_count": 7,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "len(schema)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 547
        },
        "id": "6P1_nmdtLeyq",
        "outputId": "8e332d50-b2a8-4927-e0ac-9e8dd115f433"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<Axes: xlabel='keys'>"
            ]
          },
          "execution_count": 51,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "colors = ['violet', 'indigo', 'blue', 'green', 'yellow', 'orange', 'red']\n",
        "counts = {}\n",
        "\n",
        "for tag in schema:\n",
        "  counts[tag] = 0\n",
        "  for sample in train_samples:\n",
        "    for label in sample:\n",
        "      if label[1] == tag:\n",
        "        counts[tag] += 1\n",
        "\n",
        "counts_df = pd.DataFrame({'keys': list(counts.keys()), 'values': list(counts.values())})\n",
        "counts_df.plot.bar(x='keys', y='values', legend=False, color=colors)\n",
        "# plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 529
        },
        "id": "vcCUJziAWBYX",
        "outputId": "840580ac-968c-4c7d-b79f-9bf28bba87a3"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "counts.pop('O')\n",
        "counts_df = pd.DataFrame({'keys': list(counts.keys()), 'values': list(counts.values())})\n",
        "ax = counts_df.plot.bar(x='keys', y='values', legend=False, color=colors)\n",
        "ax.bar_label(ax.containers[0])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "id": "Heu9f-z1rgFY"
      },
      "outputs": [],
      "source": [
        "from collections import namedtuple, defaultdict, Counter, OrderedDict"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "id": "K1lrrLSXQOZ7"
      },
      "outputs": [],
      "source": [
        "Tag = namedtuple(\"Tag\", [\"token\", \"tag\"])\n",
        "\n",
        "def load_sequences(filename, sep=\"\\t\", notypes=False):\n",
        "    tag_count = defaultdict(int)\n",
        "    sequences = []\n",
        "    with open(filename) as fp:\n",
        "        seq = []\n",
        "        for line in fp:\n",
        "            line = line.strip()\n",
        "            if line:\n",
        "                line = line.split(sep)\n",
        "                if notypes:\n",
        "                    line[1] = line[1][0]\n",
        "                tag_count[line[1]] += 1\n",
        "                #print line\n",
        "                seq.append(Tag(*line))\n",
        "            else:\n",
        "                sequences.append(seq)\n",
        "                seq = []\n",
        "        if seq:\n",
        "            sequences.append(seq)\n",
        "    return sequences, tag_count"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "7u6qbAaCrsBm"
      },
      "outputs": [],
      "source": [
        "sequences, tag_count = load_sequences(train_file, sep=\"\\t\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Sul1Ezi6sBVz",
        "outputId": "59fcb9d1-a038-4a64-b071-f734aa3b603d"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "2394"
            ]
          },
          "execution_count": 16,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "len(sequences)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZVWoklC8sRIy",
        "outputId": "11854594-df49-4f57-9f6e-2c37f4e47a4c"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "defaultdict(int,\n",
              "            {'O': 44007,\n",
              "             'B-geo-loc': 276,\n",
              "             'B-facility': 104,\n",
              "             'I-facility': 105,\n",
              "             'B-movie': 34,\n",
              "             'I-movie': 46,\n",
              "             'B-company': 171,\n",
              "             'B-product': 97,\n",
              "             'B-person': 449,\n",
              "             'B-other': 225,\n",
              "             'I-other': 320,\n",
              "             'B-sportsteam': 51,\n",
              "             'I-sportsteam': 23,\n",
              "             'I-product': 80,\n",
              "             'I-company': 36,\n",
              "             'I-person': 215,\n",
              "             'I-geo-loc': 49,\n",
              "             'B-tvshow': 34,\n",
              "             'B-musicartist': 55,\n",
              "             'I-musicartist': 61,\n",
              "             'I-tvshow': 31})"
            ]
          },
          "execution_count": 17,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "tag_count"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "tCYZxjSvsUX7"
      },
      "outputs": [],
      "source": [
        "df = pd.DataFrame(sum(sequences, []))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "ivdm3t3OsXsF",
        "outputId": "51e1962d-f940-4688-d499-cb0b06a1e635"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-6a0f3810-52ec-4b8f-ae09-932f9cd38058\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>token</th>\n",
              "      <th>tag</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>@SammieLynnsMom</td>\n",
              "      <td>O</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>@tg10781</td>\n",
              "      <td>O</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>they</td>\n",
              "      <td>O</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>will</td>\n",
              "      <td>O</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>be</td>\n",
              "      <td>O</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6a0f3810-52ec-4b8f-ae09-932f9cd38058')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-a5514986-dc18-446c-a8de-8e1d87b93bae\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-a5514986-dc18-446c-a8de-8e1d87b93bae')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-a5514986-dc18-446c-a8de-8e1d87b93bae button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-6a0f3810-52ec-4b8f-ae09-932f9cd38058 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-6a0f3810-52ec-4b8f-ae09-932f9cd38058');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "text/plain": [
              "             token tag\n",
              "0  @SammieLynnsMom   O\n",
              "1         @tg10781   O\n",
              "2             they   O\n",
              "3             will   O\n",
              "4               be   O"
            ]
          },
          "execution_count": 19,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hI7brNnhsY52",
        "outputId": "c934855d-6953-4181-f228-7d31dc145650"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(46469, 2)"
            ]
          },
          "execution_count": 20,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 509
        },
        "id": "ZO-Wjw7_sc1P",
        "outputId": "11547cf5-93f8-4489-fc97-aae95e825264"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "ax = df[df.tag != \"O\"][\"tag\"].value_counts().plot(kind=\"bar\")\n",
        "ax.bar_label(ax.containers[0])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "wqoTNneLsnVq"
      },
      "outputs": [],
      "source": [
        "def count_phrases(sequences, ptype=\"movie\"):\n",
        "    phrase_counts = defaultdict(int)\n",
        "    check_tag = ptype\n",
        "    for seq in sequences:\n",
        "        phrase = \"\"\n",
        "        for tag in seq:\n",
        "            if not phrase and tag.tag == \"B-%s\" % check_tag:\n",
        "                phrase = tag.token\n",
        "                continue\n",
        "            if tag.tag == \"I-%s\" % check_tag:\n",
        "                phrase += \" %s\" % tag.token\n",
        "                continue\n",
        "            if phrase:\n",
        "                phrase_counts[phrase] += 1\n",
        "                phrase = \"\"\n",
        "    return phrase_counts\n",
        "\n",
        "def count_phrases_lengths(sequences):\n",
        "    lengths = []\n",
        "    for seq in sequences:\n",
        "        phrase_len = 0\n",
        "        for tag in seq:\n",
        "            if tag.tag[0] == \"B\":\n",
        "                if phrase_len == 0:\n",
        "                    phrase_len += 1\n",
        "                else:\n",
        "                    lengths.append(phrase_len)\n",
        "                    phrase_len = 1\n",
        "                continue\n",
        "            if tag.tag[0] == \"I\":\n",
        "                phrase_len += 1\n",
        "                continue\n",
        "            if phrase_len:\n",
        "                lengths.append(phrase_len)\n",
        "                phrase_len = 0\n",
        "        if phrase_len:\n",
        "            lengths.append(phrase_len)\n",
        "            phrase_len = 0\n",
        "    return lengths"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Xq2iovnwtMKA"
      },
      "outputs": [],
      "source": [
        "lengths = count_phrases_lengths(sequences)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 453
        },
        "id": "rELfewVytRxx",
        "outputId": "e227a313-a598-4747-eb48-a2bd075a1b2a"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "ax = pd.DataFrame(sorted(Counter(lengths).items()), columns=[\"phrase_length\", \"counts\"]).plot(x=\"phrase_length\", y=\"counts\", kind=\"bar\")\n",
        "ax.bar_label(ax.containers[0])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 677
        },
        "id": "YoXlfPKXvHn5",
        "outputId": "bb492f69-75cf-4e0d-d248-92c38b959a05"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-1c588322-3bbe-4e1e-9cb6-8871eb321c8d\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>index</th>\n",
              "      <th>tag</th>\n",
              "      <th>Boundary</th>\n",
              "      <th>Class</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>B-person</td>\n",
              "      <td>449</td>\n",
              "      <td>B</td>\n",
              "      <td>person</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>I-other</td>\n",
              "      <td>320</td>\n",
              "      <td>I</td>\n",
              "      <td>other</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>B-geo-loc</td>\n",
              "      <td>276</td>\n",
              "      <td>B</td>\n",
              "      <td>geo-loc</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>B-other</td>\n",
              "      <td>225</td>\n",
              "      <td>B</td>\n",
              "      <td>other</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>I-person</td>\n",
              "      <td>215</td>\n",
              "      <td>I</td>\n",
              "      <td>person</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>B-company</td>\n",
              "      <td>171</td>\n",
              "      <td>B</td>\n",
              "      <td>company</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>I-facility</td>\n",
              "      <td>105</td>\n",
              "      <td>I</td>\n",
              "      <td>facility</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>B-facility</td>\n",
              "      <td>104</td>\n",
              "      <td>B</td>\n",
              "      <td>facility</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>B-product</td>\n",
              "      <td>97</td>\n",
              "      <td>B</td>\n",
              "      <td>product</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>I-product</td>\n",
              "      <td>80</td>\n",
              "      <td>I</td>\n",
              "      <td>product</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>I-musicartist</td>\n",
              "      <td>61</td>\n",
              "      <td>I</td>\n",
              "      <td>musicartist</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>B-musicartist</td>\n",
              "      <td>55</td>\n",
              "      <td>B</td>\n",
              "      <td>musicartist</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>B-sportsteam</td>\n",
              "      <td>51</td>\n",
              "      <td>B</td>\n",
              "      <td>sportsteam</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>I-geo-loc</td>\n",
              "      <td>49</td>\n",
              "      <td>I</td>\n",
              "      <td>geo-loc</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>I-movie</td>\n",
              "      <td>46</td>\n",
              "      <td>I</td>\n",
              "      <td>movie</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>I-company</td>\n",
              "      <td>36</td>\n",
              "      <td>I</td>\n",
              "      <td>company</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>B-tvshow</td>\n",
              "      <td>34</td>\n",
              "      <td>B</td>\n",
              "      <td>tvshow</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>B-movie</td>\n",
              "      <td>34</td>\n",
              "      <td>B</td>\n",
              "      <td>movie</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>I-tvshow</td>\n",
              "      <td>31</td>\n",
              "      <td>I</td>\n",
              "      <td>tvshow</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>I-sportsteam</td>\n",
              "      <td>23</td>\n",
              "      <td>I</td>\n",
              "      <td>sportsteam</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1c588322-3bbe-4e1e-9cb6-8871eb321c8d')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-0c8f641b-feae-4e56-aa70-3825dcf87d7c\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-0c8f641b-feae-4e56-aa70-3825dcf87d7c')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-0c8f641b-feae-4e56-aa70-3825dcf87d7c button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-1c588322-3bbe-4e1e-9cb6-8871eb321c8d button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-1c588322-3bbe-4e1e-9cb6-8871eb321c8d');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "text/plain": [
              "            index  tag Boundary        Class\n",
              "0        B-person  449        B       person\n",
              "1         I-other  320        I        other\n",
              "2       B-geo-loc  276        B      geo-loc\n",
              "3         B-other  225        B        other\n",
              "4        I-person  215        I       person\n",
              "5       B-company  171        B      company\n",
              "6      I-facility  105        I     facility\n",
              "7      B-facility  104        B     facility\n",
              "8       B-product   97        B      product\n",
              "9       I-product   80        I      product\n",
              "10  I-musicartist   61        I  musicartist\n",
              "11  B-musicartist   55        B  musicartist\n",
              "12   B-sportsteam   51        B   sportsteam\n",
              "13      I-geo-loc   49        I      geo-loc\n",
              "14        I-movie   46        I        movie\n",
              "15      I-company   36        I      company\n",
              "16       B-tvshow   34        B       tvshow\n",
              "17        B-movie   34        B        movie\n",
              "18       I-tvshow   31        I       tvshow\n",
              "19   I-sportsteam   23        I   sportsteam"
            ]
          },
          "execution_count": 36,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df_t = df[df.tag != \"O\"][\"tag\"].value_counts().reset_index()\n",
        "df_t[[\"Boundary\", \"Class\"]] = df_t[\"index\"].str.split(\"-\", 1, expand=True)\n",
        "df_t"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "WEisruSzwBN1",
        "outputId": "a9f4bb35-784f-46e9-cf90-877690978959"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1500x2000 with 10 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "grid = sns.catplot(x=\"Boundary\", y=\"tag\", col=\"Class\", data=df_t, col_wrap=3, kind=\"bar\")\n",
        "for i in range(10):\n",
        "  ax = grid.axes[i]\n",
        "  ax.bar_label(ax.containers[0])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gEz4j7m8wP82"
      },
      "outputs": [],
      "source": [
        "import regex as re"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "vbuvzDKm0Qai"
      },
      "outputs": [],
      "source": [
        "URL_REGEX = re.compile(r'^http[s]?://')\n",
        "assert URL_REGEX.match(\"http://Google.com\") is not None\n",
        "assert URL_REGEX.match(\"https://Google.com\") is not None\n",
        "assert URL_REGEX.match(\"abchttps://Google.com\") is None"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Y0GV-bak0Sl0"
      },
      "outputs": [],
      "source": [
        "REPEATED_CHAR_REGEX=re.compile(r'^[\\.\\,!\\?\"\\':;_\\-]{2,}$')\n",
        "assert REPEATED_CHAR_REGEX.match(\"!\") is None\n",
        "assert REPEATED_CHAR_REGEX.match(\"!!\") is not None\n",
        "assert REPEATED_CHAR_REGEX.match(\"!?...!\") is not None\n",
        "assert REPEATED_CHAR_REGEX.match('\".') is not None\n",
        "assert REPEATED_CHAR_REGEX.match(\"aaaaa\") is None\n",
        "assert REPEATED_CHAR_REGEX.match('\\\".') is not None"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "1FHQEOdV0VhI"
      },
      "outputs": [],
      "source": [
        "WORD_REGEX=re.compile(r'^([A-Za-z]+[\\'\\-\\.]?)+$')\n",
        "assert WORD_REGEX.match(\"ABC\") is not None\n",
        "assert WORD_REGEX.match(\"ABC-D\") is not None\n",
        "assert WORD_REGEX.match(\"ABC'D\") is not None\n",
        "assert WORD_REGEX.match(\"ABC.DFC.\") is not None\n",
        "assert WORD_REGEX.match(\"9-1\") is None\n",
        "assert WORD_REGEX.match(\"A.B.C\") is not None"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "r6P58cW80Xaw"
      },
      "outputs": [],
      "source": [
        "PUNCT_REGEX=re.compile(r'^\\p{P}$')\n",
        "assert PUNCT_REGEX.match(\".\") is not None\n",
        "assert PUNCT_REGEX.match(\"?\") is not None\n",
        "assert PUNCT_REGEX.match(\"!\") is not None\n",
        "assert PUNCT_REGEX.match(\"aa\") is None\n",
        "assert PUNCT_REGEX.match(\"a.\") is None"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "lNgtfAaJ0Z7L"
      },
      "outputs": [],
      "source": [
        "NUMBER_REGEX=re.compile(r'^(([0-9]*[.]?[0-9]+)|([0-9]+[,]?[0-9]+))$')\n",
        "assert NUMBER_REGEX.match(\"123\") is not None\n",
        "assert NUMBER_REGEX.match(\"123,4\") is not None\n",
        "assert NUMBER_REGEX.match(\"12.5\") is not None\n",
        "assert NUMBER_REGEX.match(\".5\") is not None\n",
        "assert NUMBER_REGEX.match(\"12-A\") is None\n",
        "assert NUMBER_REGEX.match(\",55\") is None\n",
        "assert NUMBER_REGEX.match(\"5.5.55\") is None"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "CBQ1UiBL0gvi"
      },
      "outputs": [],
      "source": [
        "def get_token_type(t):\n",
        "    if len(t.token) > 1 and t.token[0] in ['\\'', '\\\"', '(', '-']:\n",
        "        return get_token_type(Tag(t.token[1:], t.tag))\n",
        "    if PUNCT_REGEX.match(t.token):\n",
        "        return \"Puctuation\"\n",
        "    if len(t.token) > 1 and t.token[0] == \"@\":\n",
        "        return \"Mention\"\n",
        "    if len(t.token) > 1 and t.token[0] == \"#\":\n",
        "        return \"Hashtag\"\n",
        "    if len(t.token) > 1 and t.token[0] == \"$\":\n",
        "        return \"Money\"\n",
        "    if URL_REGEX.match(t.token):\n",
        "        return \"URL\"\n",
        "    if NUMBER_REGEX.match(t.token):\n",
        "        return \"Number\"\n",
        "    if WORD_REGEX.match(t.token):\n",
        "        return \"Word\"\n",
        "    if t.token.isalnum():\n",
        "        return \"Alphanumeric\"\n",
        "    if REPEATED_CHAR_REGEX.match(t.token):\n",
        "        return \"Repeatedchar\"\n",
        "    return \"Other\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "f9mQfOMo0jcn",
        "outputId": "5b3a138f-5fc3-4c89-e9c2-53d242505bef"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-9608f64f-d95e-4855-a5a9-3f0793e5f081\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Type</th>\n",
              "      <th>Counts</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Word</td>\n",
              "      <td>36523</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Puctuation</td>\n",
              "      <td>4848</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Mention</td>\n",
              "      <td>1300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Other</td>\n",
              "      <td>1029</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Repeatedchar</td>\n",
              "      <td>1010</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Number</td>\n",
              "      <td>657</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>Hashtag</td>\n",
              "      <td>449</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>URL</td>\n",
              "      <td>448</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>Alphanumeric</td>\n",
              "      <td>185</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>Money</td>\n",
              "      <td>20</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9608f64f-d95e-4855-a5a9-3f0793e5f081')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-05e0cca7-b801-4d72-bed5-6e8d827fc31b\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-05e0cca7-b801-4d72-bed5-6e8d827fc31b')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-05e0cca7-b801-4d72-bed5-6e8d827fc31b button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-9608f64f-d95e-4855-a5a9-3f0793e5f081 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-9608f64f-d95e-4855-a5a9-3f0793e5f081');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "text/plain": [
              "           Type  Counts\n",
              "0          Word   36523\n",
              "1    Puctuation    4848\n",
              "2       Mention    1300\n",
              "3         Other    1029\n",
              "4  Repeatedchar    1010\n",
              "5        Number     657\n",
              "6       Hashtag     449\n",
              "7           URL     448\n",
              "8  Alphanumeric     185\n",
              "9         Money      20"
            ]
          },
          "execution_count": 62,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "pd.DataFrame(Counter(get_token_type(t) for seq in sequences for t in seq).most_common(), columns=[\"Type\", \"Counts\"])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "9lGSbHMp0nA_"
      },
      "outputs": [],
      "source": [
        "other_tokens = filter(lambda x: get_token_type(x) == \"Money\", (t for seq in sequences for t in seq))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "FHycTUd60tQr"
      },
      "outputs": [],
      "source": [
        "token_counts = Counter(t.token for t in other_tokens).most_common()\n",
        "for i, t in enumerate(token_counts):\n",
        "  print(f\"[{i}] {t}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "TIloLUD006yq"
      },
      "outputs": [],
      "source": [
        "TYPE_MAPPINGS=OrderedDict({\n",
        "    \"Other\": \"OOV\",\n",
        "    \"Hashtag\": \"#Hashtag\",\n",
        "    \"Mention\": \"@Mention\",\n",
        "    \"Puctuation\": \".\",\n",
        "    \"Number\": \"1111\",\n",
        "    \"Money\": \"$100\",\n",
        "    \"URL\": \"http://URL.com\",\n",
        "    \"Repeatedchar\": \"...\"\n",
        "})\n",
        "\n",
        "def get_vocab(sequences, prepend_extras=True):\n",
        "  excluded_types=set(TYPE_MAPPINGS.keys())\n",
        "  vocab = defaultdict(int)\n",
        "  for seq in sequences:\n",
        "    for t in seq:\n",
        "      t_type = get_token_type(t)\n",
        "      if t_type not in excluded_types:\n",
        "        vocab[t.token] += 1\n",
        "  id2word = [\"OOV\"]\n",
        "  if prepend_extras:\n",
        "    id2word = [\"<MASK>\", \"<BOS>\", \"<EOS>\"] + list(TYPE_MAPPINGS.values())\n",
        "  id2word.extend([t[0] for t in sorted(vocab.items(), key=lambda x: x[1], reverse=True)])\n",
        "  print(len(id2word))\n",
        "  word2id = OrderedDict({k:i for i,k in enumerate(id2word)})\n",
        "  return id2word, word2id, vocab"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jdeFulpV23kO",
        "outputId": "7bd28fbd-2950-40f3-a269-1205193b2adf"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "8033\n"
          ]
        }
      ],
      "source": [
        "id2word, word2id, vocab = get_vocab(sequences)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "c5qdQVV73Trr"
      },
      "outputs": [],
      "source": [
        "assert len(id2word) == len(word2id)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9kHUoGIa4lLH",
        "outputId": "32b92632-755b-4e24-c029-6b34b2268cca"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[('<MASK>', 1),\n",
              " ('<BOS>', 1),\n",
              " ('<EOS>', 1),\n",
              " ('OOV', 1),\n",
              " ('#Hashtag', 1),\n",
              " ('@Mention', 1),\n",
              " ('.', 1),\n",
              " ('1111', 1),\n",
              " ('$100', 1),\n",
              " ('http://URL.com', 1)]"
            ]
          },
          "execution_count": 100,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "Counter(id2word).most_common(10)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H3yBG6s14pfv",
        "outputId": "24377a80-d1e7-4306-8723-62e6b3d1e0a7"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Done\n",
            "8023\n",
            "Done\n"
          ]
        }
      ],
      "source": [
        "with open(\"vocab.txt\", \"w+\") as fp:\n",
        "    for t in id2word:\n",
        "        print(t, file=fp)\n",
        "print(\"Done\")\n",
        "\n",
        "id2word, word2id, vocab = get_vocab(sequences, prepend_extras=False)\n",
        "with open(\"vocab.no_extras.txt\", \"w+\") as fp:\n",
        "    for t in id2word:\n",
        "        print(t, file=fp)\n",
        "print(\"Done\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "07os1Zpy4_zx"
      },
      "outputs": [],
      "source": [
        "def phrases_per_seq(sequences):\n",
        "    count_phrases = []\n",
        "    count_phrase_types = defaultdict(list)\n",
        "    for seq in sequences:\n",
        "        current_phrase_count = 0\n",
        "        phrase_type_count = defaultdict(int)\n",
        "        for t in seq:\n",
        "            if t.tag[0] in (\"U\", \"B\"):\n",
        "                current_phrase_count += 1\n",
        "                phrase_type_count[t.tag[2:]] += 1\n",
        "        count_phrases.append(current_phrase_count)\n",
        "        for k,v in phrase_type_count.items():\n",
        "            count_phrase_types[k].append(v)\n",
        "    return count_phrases, count_phrase_types"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "mYtIH8zQ5TF0"
      },
      "outputs": [],
      "source": [
        "count_phrases, count_phrase_types = phrases_per_seq(sequences)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        },
        "id": "d4VrgydN5Vl-",
        "outputId": "2c206d05-f6f9-4032-e165-e23a5fc86028"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjAAAAGdCAYAAAAMm0nCAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAloUlEQVR4nO3df3BU9b3/8dcmIT+K7IaEm112TCC34whBikowrii3XjIEiNxhmtaba2pz2wy515ugEEWSa4m/DcRWMYBQHK8wYxit00I1jtHc0Ev8EUIITYEI0TtFiDKb2AnZNXFIQrLfP7ycb1exgN24+STPx8yZ6Z7z2d33unX2OSe7R1sgEAgIAADAIBHhHgAAAOByETAAAMA4BAwAADAOAQMAAIxDwAAAAOMQMAAAwDgEDAAAMA4BAwAAjBMV7gFGyvDwsE6fPq1JkybJZrOFexwAAHAJAoGAPvvsM7ndbkVEfP15ljEbMKdPn1ZycnK4xwAAAN9AR0eHrrzyyq89PmYDZtKkSZK++Adgt9vDPA0AALgUfr9fycnJ1uf41xmzAXP+z0Z2u52AAQDAMBf7+gdf4gUAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHGiwj2AiaaXvh7uEULio/XZ4R4BAIBvhDMwAADAOAQMAAAwDgEDAACMQ8AAAADjEDAAAMA4BAwAADAOAQMAAIxDwAAAAOMQMAAAwDgEDAAAMA4BAwAAjEPAAAAA4xAwAADAOJcdMA0NDVq2bJncbrdsNpv27NnztWv//d//XTabTRs3bgza393drby8PNntdsXHx6ugoEC9vb1Baw4fPqxbbrlFsbGxSk5OVmVl5eWOCgAAxqjLDpi+vj7NmTNHW7Zs+avrdu/erf3798vtdn/lWF5entra2lRXV6eamho1NDSosLDQOu73+7Vo0SJNmzZNLS0tevLJJ/XQQw9p+/btlzsuAAAYg6Iu9w5LlizRkiVL/uqaTz75RCtXrtSbb76p7OzsoGPHjh1TbW2tmpublZ6eLknatGmTli5dql/84hdyu92qrq7WwMCA/uu//kvR0dGaNWuWWltb9dRTTwWFDgAAGJ9C/h2Y4eFh3XnnnVqzZo1mzZr1leONjY2Kj4+34kWSMjMzFRERoaamJmvNggULFB0dba3JyspSe3u7zpw5c8Hn7e/vl9/vD9oAAMDYFPKA2bBhg6KionT33Xdf8LjX61VSUlLQvqioKCUkJMjr9VprnE5n0Jrzt8+v+bKKigo5HA5rS05O/ltfCgAAGKVCGjAtLS165plntGPHDtlstlA+9EWVlZXJ5/NZW0dHx7f6/AAA4NsT0oB5++231dXVpZSUFEVFRSkqKkonT57Uvffeq+nTp0uSXC6Xurq6gu537tw5dXd3y+VyWWs6OzuD1py/fX7Nl8XExMhutwdtAABgbAppwNx55506fPiwWltbrc3tdmvNmjV68803JUkej0c9PT1qaWmx7rd3714NDw8rIyPDWtPQ0KDBwUFrTV1dna6++mpNnjw5lCMDAAADXfavkHp7e/W///u/1u0TJ06otbVVCQkJSklJUWJiYtD6CRMmyOVy6eqrr5YkzZw5U4sXL9aKFSu0bds2DQ4Oqri4WLm5udZPru+44w49/PDDKigo0Nq1a3X06FE988wzevrpp/+W1woAAMaIyw6YgwcP6tZbb7Vul5SUSJLy8/O1Y8eOS3qM6upqFRcXa+HChYqIiFBOTo6qqqqs4w6HQ2+99ZaKioo0d+5cTZkyReXl5fyEGgAASJJsgUAgEO4hRoLf75fD4ZDP5wv592Gml74e0scLl4/WZ198EQAA36JL/fzmv4UEAACMQ8AAAADjEDAAAMA4BAwAADAOAQMAAIxDwAAAAOMQMAAAwDgEDAAAMA4BAwAAjEPAAAAA4xAwAADAOAQMAAAwDgEDAACMQ8AAAADjEDAAAMA4BAwAADAOAQMAAIxDwAAAAOMQMAAAwDgEDAAAMA4BAwAAjEPAAAAA4xAwAADAOAQMAAAwDgEDAACMQ8AAAADjEDAAAMA4BAwAADAOAQMAAIxDwAAAAOMQMAAAwDgEDAAAMA4BAwAAjEPAAAAA4xAwAADAOAQMAAAwDgEDAACMc9kB09DQoGXLlsntdstms2nPnj3WscHBQa1du1azZ8/WxIkT5Xa79ZOf/ESnT58Oeozu7m7l5eXJbrcrPj5eBQUF6u3tDVpz+PBh3XLLLYqNjVVycrIqKyu/2SsEAABjzmUHTF9fn+bMmaMtW7Z85djnn3+uQ4cOad26dTp06JB++9vfqr29Xf/0T/8UtC4vL09tbW2qq6tTTU2NGhoaVFhYaB33+/1atGiRpk2bppaWFj355JN66KGHtH379m/wEgEAwFhjCwQCgW98Z5tNu3fv1vLly792TXNzs2644QadPHlSKSkpOnbsmNLS0tTc3Kz09HRJUm1trZYuXaqPP/5YbrdbW7du1QMPPCCv16vo6GhJUmlpqfbs2aPjx49f0mx+v18Oh0M+n092u/2bvsQLml76ekgfL1w+Wp8d7hEAAAhyqZ/fI/4dGJ/PJ5vNpvj4eElSY2Oj4uPjrXiRpMzMTEVERKipqclas2DBAiteJCkrK0vt7e06c+bMBZ+nv79ffr8/aAMAAGPTiAbM2bNntXbtWv3Lv/yLVVFer1dJSUlB66KiopSQkCCv12utcTqdQWvO3z6/5ssqKirkcDisLTk5OdQvBwAAjBIjFjCDg4O6/fbbFQgEtHXr1pF6GktZWZl8Pp+1dXR0jPhzAgCA8IgaiQc9Hy8nT57U3r17g/6G5XK51NXVFbT+3Llz6u7ulsvlstZ0dnYGrTl/+/yaL4uJiVFMTEwoXwYAABilQn4G5ny8fPjhh/rv//5vJSYmBh33eDzq6elRS0uLtW/v3r0aHh5WRkaGtaahoUGDg4PWmrq6Ol199dWaPHlyqEcGAACGueyA6e3tVWtrq1pbWyVJJ06cUGtrq06dOqXBwUH98Ic/1MGDB1VdXa2hoSF5vV55vV4NDAxIkmbOnKnFixdrxYoVOnDggN59910VFxcrNzdXbrdbknTHHXcoOjpaBQUFamtr08svv6xnnnlGJSUloXvlAADAWJf9M+r/+Z//0a233vqV/fn5+XrooYeUmpp6wfv9/ve/1/e//31JX1zIrri4WK+99poiIiKUk5OjqqoqXXHFFdb6w4cPq6ioSM3NzZoyZYpWrlyptWvXXvKc/Iz64vgZNQBgtLnUz++/6TowoxkBc3EEDABgtBk114EBAAAINQIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgnMsOmIaGBi1btkxut1s2m0179uwJOh4IBFReXq6pU6cqLi5OmZmZ+vDDD4PWdHd3Ky8vT3a7XfHx8SooKFBvb2/QmsOHD+uWW25RbGyskpOTVVlZefmvDgAAjEmXHTB9fX2aM2eOtmzZcsHjlZWVqqqq0rZt29TU1KSJEycqKytLZ8+etdbk5eWpra1NdXV1qqmpUUNDgwoLC63jfr9fixYt0rRp09TS0qInn3xSDz30kLZv3/4NXiIAABhrbIFAIPCN72yzaffu3Vq+fLmkL86+uN1u3XvvvbrvvvskST6fT06nUzt27FBubq6OHTumtLQ0NTc3Kz09XZJUW1urpUuX6uOPP5bb7dbWrVv1wAMPyOv1Kjo6WpJUWlqqPXv26Pjx45c0m9/vl8PhkM/nk91u/6Yv8YKml74e0scLl4/WZ4d7BAAAglzq53dIvwNz4sQJeb1eZWZmWvscDocyMjLU2NgoSWpsbFR8fLwVL5KUmZmpiIgINTU1WWsWLFhgxYskZWVlqb29XWfOnLngc/f398vv9wdtAABgbAppwHi9XkmS0+kM2u90Oq1jXq9XSUlJQcejoqKUkJAQtOZCj/GXz/FlFRUVcjgc1pacnPy3vyAAADAqjZlfIZWVlcnn81lbR0dHuEcCAAAjJKQB43K5JEmdnZ1B+zs7O61jLpdLXV1dQcfPnTun7u7uoDUXeoy/fI4vi4mJkd1uD9oAAMDYFNKASU1NlcvlUn19vbXP7/erqalJHo9HkuTxeNTT06OWlhZrzd69ezU8PKyMjAxrTUNDgwYHB601dXV1uvrqqzV58uRQjgwAAAx02QHT29ur1tZWtba2Svrii7utra06deqUbDabVq1apccee0yvvvqqjhw5op/85Cdyu93WL5VmzpypxYsXa8WKFTpw4IDeffddFRcXKzc3V263W5J0xx13KDo6WgUFBWpra9PLL7+sZ555RiUlJSF74QAAwFxRl3uHgwcP6tZbb7Vun4+K/Px87dixQ/fff7/6+vpUWFionp4e3XzzzaqtrVVsbKx1n+rqahUXF2vhwoWKiIhQTk6OqqqqrOMOh0NvvfWWioqKNHfuXE2ZMkXl5eVB14oBAADj1990HZjRjOvAXBzXgQEAjDZhuQ4MAADAt4GAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYJyQB8zQ0JDWrVun1NRUxcXF6bvf/a4effRRBQIBa00gEFB5ebmmTp2quLg4ZWZm6sMPPwx6nO7ubuXl5clutys+Pl4FBQXq7e0N9bgAAMBAIQ+YDRs2aOvWrdq8ebOOHTumDRs2qLKyUps2bbLWVFZWqqqqStu2bVNTU5MmTpyorKwsnT171lqTl5entrY21dXVqaamRg0NDSosLAz1uAAAwEC2wF+eGgmB2267TU6nU88//7y1LycnR3FxcXrxxRcVCATkdrt177336r777pMk+Xw+OZ1O7dixQ7m5uTp27JjS0tLU3Nys9PR0SVJtba2WLl2qjz/+WG63+6Jz+P1+ORwO+Xw+2e32UL5ETS99PaSPFy4frc8O9wgAAAS51M/vkJ+Buemmm1RfX68PPvhAkvTHP/5R77zzjpYsWSJJOnHihLxerzIzM637OBwOZWRkqLGxUZLU2Nio+Ph4K14kKTMzUxEREWpqarrg8/b398vv9wdtAABgbIoK9QOWlpbK7/drxowZioyM1NDQkB5//HHl5eVJkrxeryTJ6XQG3c/pdFrHvF6vkpKSggeNilJCQoK15ssqKir08MMPh/rlAACAUSjkZ2B+/etfq7q6Wrt27dKhQ4e0c+dO/eIXv9DOnTtD/VRBysrK5PP5rK2jo2NEnw8AAIRPyM/ArFmzRqWlpcrNzZUkzZ49WydPnlRFRYXy8/PlcrkkSZ2dnZo6dap1v87OTl177bWSJJfLpa6urqDHPXfunLq7u637f1lMTIxiYmJC/XIAAMAoFPIzMJ9//rkiIoIfNjIyUsPDw5Kk1NRUuVwu1dfXW8f9fr+amprk8XgkSR6PRz09PWppabHW7N27V8PDw8rIyAj1yAAAwDAhPwOzbNkyPf7440pJSdGsWbP0hz/8QU899ZR+9rOfSZJsNptWrVqlxx57TFdddZVSU1O1bt06ud1uLV++XJI0c+ZMLV68WCtWrNC2bds0ODio4uJi5ebmXtIvkAAAwNgW8oDZtGmT1q1bp//4j/9QV1eX3G63/u3f/k3l5eXWmvvvv199fX0qLCxUT0+Pbr75ZtXW1io2NtZaU11dreLiYi1cuFARERHKyclRVVVVqMcFAAAGCvl1YEYLrgNzcVwHBgAw2oTtOjAAAAAjjYABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYZ0QC5pNPPtGPf/xjJSYmKi4uTrNnz9bBgwet44FAQOXl5Zo6dari4uKUmZmpDz/8MOgxuru7lZeXJ7vdrvj4eBUUFKi3t3ckxgUAAIYJecCcOXNG8+fP14QJE/TGG2/o/fff1y9/+UtNnjzZWlNZWamqqipt27ZNTU1NmjhxorKysnT27FlrTV5entra2lRXV6eamho1NDSosLAw1OMCAAAD2QKBQCCUD1haWqp3331Xb7/99gWPBwIBud1u3XvvvbrvvvskST6fT06nUzt27FBubq6OHTumtLQ0NTc3Kz09XZJUW1urpUuX6uOPP5bb7b7oHH6/Xw6HQz6fT3a7PXQvUNL00tdD+njh8tH67HCPAABAkEv9/A75GZhXX31V6enp+tGPfqSkpCRdd911eu6556zjJ06ckNfrVWZmprXP4XAoIyNDjY2NkqTGxkbFx8db8SJJmZmZioiIUFNT0wWft7+/X36/P2gDAABjU8gD5k9/+pO2bt2qq666Sm+++abuuusu3X333dq5c6ckyev1SpKcTmfQ/ZxOp3XM6/UqKSkp6HhUVJQSEhKsNV9WUVEhh8NhbcnJyaF+aQAAYJQIecAMDw/r+uuv1xNPPKHrrrtOhYWFWrFihbZt2xbqpwpSVlYmn89nbR0dHSP6fAAAIHxCHjBTp05VWlpa0L6ZM2fq1KlTkiSXyyVJ6uzsDFrT2dlpHXO5XOrq6go6fu7cOXV3d1trviwmJkZ2uz1oAwAAY1PIA2b+/Plqb28P2vfBBx9o2rRpkqTU1FS5XC7V19dbx/1+v5qamuTxeCRJHo9HPT09amlpsdbs3btXw8PDysjICPXIAADAMFGhfsDVq1frpptu0hNPPKHbb79dBw4c0Pbt27V9+3ZJks1m06pVq/TYY4/pqquuUmpqqtatWye3263ly5dL+uKMzeLFi60/PQ0ODqq4uFi5ubmX9AskAAAwtoU8YObNm6fdu3errKxMjzzyiFJTU7Vx40bl5eVZa+6//3719fWpsLBQPT09uvnmm1VbW6vY2FhrTXV1tYqLi7Vw4UJFREQoJydHVVVVoR4XAAAYKOTXgRktuA7MxXEdGADAaBO268AAAACMNAIGAAAYh4ABAADGIWAAAIBxCBgAAGCckP+MGuYYK7+mkvhFFQCMN5yBAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGCcEQ+Y9evXy2azadWqVda+s2fPqqioSImJibriiiuUk5Ojzs7OoPudOnVK2dnZ+s53vqOkpCStWbNG586dG+lxAQCAAUY0YJqbm/WrX/1K3/ve94L2r169Wq+99ppeeeUV7du3T6dPn9YPfvAD6/jQ0JCys7M1MDCg9957Tzt37tSOHTtUXl4+kuMCAABDjFjA9Pb2Ki8vT88995wmT55s7ff5fHr++ef11FNP6R//8R81d+5cvfDCC3rvvfe0f/9+SdJbb72l999/Xy+++KKuvfZaLVmyRI8++qi2bNmigYGBkRoZAAAYYsQCpqioSNnZ2crMzAza39LSosHBwaD9M2bMUEpKihobGyVJjY2Nmj17tpxOp7UmKytLfr9fbW1tF3y+/v5++f3+oA0AAIxNUSPxoC+99JIOHTqk5ubmrxzzer2Kjo5WfHx80H6n0ymv12ut+ct4OX/8/LELqaio0MMPPxyC6QEAwGgX8jMwHR0duueee1RdXa3Y2NhQP/zXKisrk8/ns7aOjo5v7bkBAMC3K+QB09LSoq6uLl1//fWKiopSVFSU9u3bp6qqKkVFRcnpdGpgYEA9PT1B9+vs7JTL5ZIkuVyur/wq6fzt82u+LCYmRna7PWgDAABjU8gDZuHChTpy5IhaW1utLT09XXl5edb/njBhgurr6637tLe369SpU/J4PJIkj8ejI0eOqKury1pTV1cnu92utLS0UI8MAAAME/LvwEyaNEnXXHNN0L6JEycqMTHR2l9QUKCSkhIlJCTIbrdr5cqV8ng8uvHGGyVJixYtUlpamu68805VVlbK6/Xq5z//uYqKihQTExPqkQEAgGFG5Eu8F/P0008rIiJCOTk56u/vV1ZWlp599lnreGRkpGpqanTXXXfJ4/Fo4sSJys/P1yOPPBKOcQEAwChjCwQCgXAPMRL8fr8cDod8Pl/Ivw8zvfT1kD4e/nYfrc8O9wgAgBC41M9v/ltIAADAOAQMAAAwDgEDAACMQ8AAAADjEDAAAMA4BAwAADAOAQMAAIxDwAAAAOMQMAAAwDgEDAAAMA4BAwAAjEPAAAAA4xAwAADAOAQMAAAwDgEDAACMQ8AAAADjEDAAAMA4BAwAADAOAQMAAIxDwAAAAOMQMAAAwDgEDAAAMA4BAwAAjEPAAAAA4xAwAADAOAQMAAAwDgEDAACMQ8AAAADjEDAAAMA4BAwAADAOAQMAAIxDwAAAAOMQMAAAwDhR4R4ACIXppa+He4SQ+Gh9drhHAAAjcAYGAAAYh4ABAADGCXnAVFRUaN68eZo0aZKSkpK0fPlytbe3B605e/asioqKlJiYqCuuuEI5OTnq7OwMWnPq1CllZ2frO9/5jpKSkrRmzRqdO3cu1OMCAAADhTxg9u3bp6KiIu3fv191dXUaHBzUokWL1NfXZ61ZvXq1XnvtNb3yyivat2+fTp8+rR/84AfW8aGhIWVnZ2tgYEDvvfeedu7cqR07dqi8vDzU4wIAAAPZAoFAYCSf4NNPP1VSUpL27dunBQsWyOfz6e/+7u+0a9cu/fCHP5QkHT9+XDNnzlRjY6NuvPFGvfHGG7rtttt0+vRpOZ1OSdK2bdu0du1affrpp4qOjr7o8/r9fjkcDvl8Ptnt9pC+prHyhVGMPnyJF8B4d6mf3yP+HRifzydJSkhIkCS1tLRocHBQmZmZ1poZM2YoJSVFjY2NkqTGxkbNnj3bihdJysrKkt/vV1tb20iPDAAARrkR/Rn18PCwVq1apfnz5+uaa66RJHm9XkVHRys+Pj5ordPplNfrtdb8ZbycP37+2IX09/erv7/fuu33+0P1MgAAwCgzomdgioqKdPToUb300ksj+TSSvvjysMPhsLbk5OQRf04AABAeIxYwxcXFqqmp0e9//3tdeeWV1n6Xy6WBgQH19PQEre/s7JTL5bLWfPlXSedvn1/zZWVlZfL5fNbW0dERwlcDAABGk5AHTCAQUHFxsXbv3q29e/cqNTU16PjcuXM1YcIE1dfXW/va29t16tQpeTweSZLH49GRI0fU1dVlramrq5PdbldaWtoFnzcmJkZ2uz1oAwAAY1PIvwNTVFSkXbt26Xe/+50mTZpkfWfF4XAoLi5ODodDBQUFKikpUUJCgux2u1auXCmPx6Mbb7xRkrRo0SKlpaXpzjvvVGVlpbxer37+85+rqKhIMTExoR4ZAAAYJuQBs3XrVknS97///aD9L7zwgv71X/9VkvT0008rIiJCOTk56u/vV1ZWlp599llrbWRkpGpqanTXXXfJ4/Fo4sSJys/P1yOPPBLqcQEAgIFG/Dow4cJ1YGAirgMDYLwbNdeBAQAACDUCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGiQr3AAD+v+mlr4d7hJD4aH12uEcAMMZxBgYAABiHgAEAAMYhYAAAgHEIGAAAYBwCBgAAGIeAAQAAxiFgAACAcQgYAABgHAIGAAAYh4ABAADGIWAAAIBxCBgAAGAcAgYAABiHgAEAAMYhYAAAgHGiwj0AgLFneunr4R4hJD5anx3uEQB8Dc7AAAAA44zqgNmyZYumT5+u2NhYZWRk6MCBA+EeCQAAjAKjNmBefvlllZSU6MEHH9ShQ4c0Z84cZWVlqaurK9yjAQCAMLMFAoFAuIe4kIyMDM2bN0+bN2+WJA0PDys5OVkrV65UaWnpRe/v9/vlcDjk8/lkt9tDOttY+fs+gPGB7/LAJJf6+T0qv8Q7MDCglpYWlZWVWfsiIiKUmZmpxsbGC96nv79f/f391m2fzyfpi38QoTbc/3nIHxMARkrK6lfCPUJIHH04K9wj4Ftw/nP7YudXRmXA/PnPf9bQ0JCcTmfQfqfTqePHj1/wPhUVFXr44Ye/sj85OXlEZgQAfLscG8M9Ab5Nn332mRwOx9ceH5UB802UlZWppKTEuj08PKzu7m4lJibKZrOF7Hn8fr+Sk5PV0dER8j9N4ZvhPRldeD9GF96P0YX34+ICgYA+++wzud3uv7puVAbMlClTFBkZqc7OzqD9nZ2dcrlcF7xPTEyMYmJigvbFx8eP1Iiy2+38n2+U4T0ZXXg/Rhfej9GF9+Ov+2tnXs4blb9Cio6O1ty5c1VfX2/tGx4eVn19vTweTxgnAwAAo8GoPAMjSSUlJcrPz1d6erpuuOEGbdy4UX19ffrpT38a7tEAAECYjdqA+ed//md9+umnKi8vl9fr1bXXXqva2tqvfLH32xYTE6MHH3zwK3+uQvjwnowuvB+jC+/H6ML7ETqj9jowAAAAX2dUfgcGAADgryFgAACAcQgYAABgHAIGAAAYh4C5TFu2bNH06dMVGxurjIwMHThwINwjjUsVFRWaN2+eJk2apKSkJC1fvlzt7e3hHgv/Z/369bLZbFq1alW4RxnXPvnkE/34xz9WYmKi4uLiNHv2bB08eDDcY41LQ0NDWrdunVJTUxUXF6fvfve7evTRRy/63/vB1yNgLsPLL7+skpISPfjggzp06JDmzJmjrKwsdXV1hXu0cWffvn0qKirS/v37VVdXp8HBQS1atEh9fX3hHm3ca25u1q9+9St973vfC/co49qZM2c0f/58TZgwQW+88Ybef/99/fKXv9TkyZPDPdq4tGHDBm3dulWbN2/WsWPHtGHDBlVWVmrTpk3hHs1Y/Iz6MmRkZGjevHnavHmzpC+uDpycnKyVK1eqtLQ0zNONb59++qmSkpK0b98+LViwINzjjFu9vb26/vrr9eyzz+qxxx7Ttddeq40bN4Z7rHGptLRU7777rt5+++1wjwJJt912m5xOp55//nlrX05OjuLi4vTiiy+GcTJzcQbmEg0MDKilpUWZmZnWvoiICGVmZqqxsTGMk0GSfD6fJCkhISHMk4xvRUVFys7ODvr3BOHx6quvKj09XT/60Y+UlJSk6667Ts8991y4xxq3brrpJtXX1+uDDz6QJP3xj3/UO++8oyVLloR5MnON2ivxjjZ//vOfNTQ09JUrATudTh0/fjxMU0H64kzYqlWrNH/+fF1zzTXhHmfceumll3To0CE1NzeHexRI+tOf/qStW7eqpKRE//mf/6nm5mbdfffdio6OVn5+frjHG3dKS0vl9/s1Y8YMRUZGamhoSI8//rjy8vLCPZqxCBgYr6ioSEePHtU777wT7lHGrY6ODt1zzz2qq6tTbGxsuMeBvgj79PR0PfHEE5Kk6667TkePHtW2bdsImDD49a9/rerqau3atUuzZs1Sa2urVq1aJbfbzfvxDREwl2jKlCmKjIxUZ2dn0P7Ozk65XK4wTYXi4mLV1NSooaFBV155ZbjHGbdaWlrU1dWl66+/3to3NDSkhoYGbd68Wf39/YqMjAzjhOPP1KlTlZaWFrRv5syZ+s1vfhOmica3NWvWqLS0VLm5uZKk2bNn6+TJk6qoqCBgviG+A3OJoqOjNXfuXNXX11v7hoeHVV9fL4/HE8bJxqdAIKDi4mLt3r1be/fuVWpqarhHGtcWLlyoI0eOqLW11drS09OVl5en1tZW4iUM5s+f/5VLC3zwwQeaNm1amCYa3z7//HNFRAR/5EZGRmp4eDhME5mPMzCXoaSkRPn5+UpPT9cNN9ygjRs3qq+vTz/96U/DPdq4U1RUpF27dul3v/udJk2aJK/XK0lyOByKi4sL83Tjz6RJk77y/aOJEycqMTGR7yWFyerVq3XTTTfpiSee0O23364DBw5o+/bt2r59e7hHG5eWLVumxx9/XCkpKZo1a5b+8Ic/6KmnntLPfvazcI9mrgAuy6ZNmwIpKSmB6OjowA033BDYv39/uEcalyRdcHvhhRfCPRr+zz/8wz8E7rnnnnCPMa699tprgWuuuSYQExMTmDFjRmD79u3hHmnc8vv9gXvuuSeQkpISiI2NDfz93/994IEHHgj09/eHezRjcR0YAABgHL4DAwAAjEPAAAAA4xAwAADAOAQMAAAwDgEDAACMQ8AAAADjEDAAAMA4BAwAADAOAQMAAIxDwAAAAOMQMAAAwDgEDAAAMM7/A8r7cjGCBRYTAAAAAElFTkSuQmCC",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.hist(count_phrases, bins=range(10))\n",
        "# ax.bar_label(ax.containers[0])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YmniE4hN8I89",
        "outputId": "3c406b0d-7e4f-4055-ac0c-1b037015359d"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "2394"
            ]
          },
          "execution_count": 129,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "len(count_phrases)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 507
        },
        "id": "CN5qpZ9b5cgp",
        "outputId": "c40dfb94-7d55-4ed5-981c-92cca32fafce"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(8, 5))\n",
        "\n",
        "num_bins = 20  # Adjust the number of bins as needed\n",
        "\n",
        "n, bins, patches = plt.hist(count_phrases, bins=num_bins, edgecolor='black')\n",
        "\n",
        "# Add bar labels\n",
        "for i in range(len(patches)):\n",
        "    height = int(n[i])\n",
        "    if height > 0:\n",
        "        plt.text(patches[i].get_x() + patches[i].get_width() / 2, height,\n",
        "                 str(height), ha='center', va='bottom')\n",
        "\n",
        "plt.xlabel('Counts')\n",
        "plt.ylabel('Frequency')\n",
        "plt.title('Histogram of Count Phrases')\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VNbvXMQz6Ypq",
        "outputId": "dad481b8-1577-4b40-fd78-afb1cf94000b"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Counter({0: 1476, 1: 549, 2: 231, 3: 92, 4: 28, 5: 11, 6: 7})"
            ]
          },
          "execution_count": 137,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "Counter(sorted(count_phrases))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "waWH7x4x9I3n",
        "outputId": "b16d78a9-cf9f-43e7-ded4-bea95e18c502"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "61.65413533834587"
            ]
          },
          "execution_count": 138,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "Counter(count_phrases)[0] * 100. / len(sequences)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "l5ZNPQ8L9auK",
        "outputId": "a2aee12e-a24b-4505-aa15-53ac3359161a"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "445"
            ]
          },
          "execution_count": 139,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "count_phrases.index(6)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6q3gBSgc9e02",
        "outputId": "de3f8e98-2fbc-4924-e931-78272064a074"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[Tag(token='Yahoo', tag='B-company'),\n",
              " Tag(token=':', tag='O'),\n",
              " Tag(token='Jackie', tag='B-person'),\n",
              " Tag(token='joins', tag='O'),\n",
              " Tag(token='Twitter', tag='B-company'),\n",
              " Tag(token='50', tag='O'),\n",
              " Tag(token='years', tag='O'),\n",
              " Tag(token='after', tag='O'),\n",
              " Tag(token='JFK', tag='B-person'),\n",
              " Tag(token='campaign', tag='O'),\n",
              " Tag(token=':', tag='O'),\n",
              " Tag(token='Had', tag='O'),\n",
              " Tag(token='Twitter', tag='B-company'),\n",
              " Tag(token='been', tag='O'),\n",
              " Tag(token='an', tag='O'),\n",
              " Tag(token='option', tag='O'),\n",
              " Tag(token='in', tag='O'),\n",
              " Tag(token='1960', tag='O'),\n",
              " Tag(token=',', tag='O'),\n",
              " Tag(token='a', tag='O'),\n",
              " Tag(token='pregnant', tag='O'),\n",
              " Tag(token='Jacquelin', tag='B-person'),\n",
              " Tag(token='...', tag='O'),\n",
              " Tag(token='http://yhoo.it/9VAD4A', tag='O')]"
            ]
          },
          "execution_count": 140,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "sequences[445]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nL5097Sf9iRg",
        "outputId": "380ab258-8134-4c7f-eb8f-6f3fc45d6c4d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "geo-loc Counter({1: 162, 2: 43, 4: 4, 3: 4})\n",
            "facility Counter({1: 84, 2: 10})\n",
            "movie Counter({1: 20, 3: 2, 2: 2, 4: 1})\n",
            "company Counter({1: 134, 2: 14, 3: 3})\n",
            "product Counter({1: 56, 2: 12, 3: 3, 4: 2})\n",
            "person Counter({1: 269, 2: 54, 3: 17, 4: 4, 5: 1})\n",
            "other Counter({1: 166, 2: 23, 3: 3, 4: 1})\n",
            "sportsteam Counter({1: 33, 2: 9})\n",
            "tvshow Counter({1: 32, 2: 1})\n",
            "musicartist Counter({1: 35, 3: 4, 2: 4})\n"
          ]
        }
      ],
      "source": [
        "for k, v in count_phrase_types.items():\n",
        "  print(k, Counter(v))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "fx46B3Gv9paW"
      },
      "outputs": [],
      "source": [
        "def phrases_context_words(sequences, lowercase=False):\n",
        "    context_words = defaultdict(lambda : defaultdict(int))\n",
        "    global_word_counts = defaultdict(int)\n",
        "    for seq in sequences:\n",
        "        phrase_types = set()\n",
        "        outside_word_counts = defaultdict(int)\n",
        "        for t in seq:\n",
        "            token, tag = t\n",
        "            if lowercase:\n",
        "                token = token.lower()\n",
        "            if tag[0] in (\"U\", \"B\"):\n",
        "                phrase_types.add(tag[2:])\n",
        "            elif t.tag[0] == \"O\":\n",
        "                outside_word_counts[token] += 1\n",
        "        for k,v in outside_word_counts.items():\n",
        "            global_word_counts[k] += 1\n",
        "            for p in phrase_types:\n",
        "                context_words[p][k] += v\n",
        "    return context_words, global_word_counts"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ucpmOLu2DxeW"
      },
      "outputs": [],
      "source": [
        "context_words, global_word_counts = phrases_context_words(sequences, lowercase=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aq4ZjTBCD0Pl",
        "outputId": "7fcb4f14-af72-49b5-f693-05f5df554db7"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "dict_keys(['facility', 'geo-loc', 'movie', 'company', 'product', 'other', 'person', 'sportsteam', 'tvshow', 'musicartist'])"
            ]
          },
          "execution_count": 30,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "context_words.keys()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "RqvY5qp2FK82"
      },
      "outputs": [],
      "source": [
        "def get_diff_type_occurence(w,context_words):\n",
        "    n = 0\n",
        "    for k, v in context_words.items():\n",
        "        if w in v:\n",
        "            n += 1\n",
        "    return n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O1BjMbAJFRYD",
        "outputId": "d8518235-d73b-4fe5-d10c-436e51f31382"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Top 10 tokens in category facility: \n",
            "\tnight: 10\n",
            "\tsoyrizo: 2\n",
            "\ttamales: 2\n",
            "\tsalsa: 2\n",
            "\tsafety: 2\n",
            "\tself-defense: 2\n",
            "\t@paulwalk: 2\n",
            "\tview: 2\n",
            "\tsaturday: 5\n",
            "\twhere: 6\n",
            "Top 10 tokens in category geo-loc: \n",
            "\tparty: 10\n",
            "\tfollow: 6\n",
            "\tf: 3\n",
            "\tsony/bud: 3\n",
            "\tlight: 3\n",
            "\t3d: 3\n",
            "\thttp://bit.ly/cds8pd: 3\n",
            "\tforward: 3\n",
            "\tbeautiful: 4\n",
            "\tetc: 4\n",
            "Top 10 tokens in category movie: \n",
            "\tmovie: 5\n",
            "\t@glasgowfilm: 2\n",
            "\tapplications: 2\n",
            "\thttp://bit.ly/bhulum: 2\n",
            "\t5:00: 2\n",
            "\t7:00: 2\n",
            "\t8:45: 2\n",
            "\tmovies: 3\n",
            "\treally: 5\n",
            "\t:p: 2\n",
            "Top 10 tokens in category company: \n",
            "\tdrama: 7\n",
            "\t@liltwist: 5\n",
            "\tedition: 7\n",
            "\tvideos: 9\n",
            "\tmeltdowns: 5\n",
            "\tepic: 6\n",
            "\tcryin: 3\n",
            "\thttp://bit.ly/cuy0yt: 3\n",
            "\tusually: 4\n",
            "\temail: 4\n",
            "Top 10 tokens in category product: \n",
            "\tmascara: 2\n",
            "\ttank: 2\n",
            "\trecord: 2\n",
            "\trefresh: 2\n",
            "\thand: 2\n",
            "\tmonday: 3\n",
            "\twant: 5\n",
            "\t@corsairartisan: 2\n",
            "\tarrived: 2\n",
            "\tships: 2\n",
            "Top 10 tokens in category other: \n",
            "\tstressful: 3\n",
            "\tstyle: 3\n",
            "\tdriving: 7\n",
            "\tnap: 3\n",
            "\tthedeandreway: 3\n",
            "\t@souljaboy: 3\n",
            "\tlist: 4\n",
            "\tnov: 4\n",
            "\tmonday: 4\n",
            "\taway: 4\n",
            "Top 10 tokens in category person: \n",
            "\t...\": 14\n",
            "\ttook: 17\n",
            "\tpart: 14\n",
            "\twake: 7\n",
            "\tconcert: 10\n",
            "\ttrampled: 6\n",
            "\tyoiu: 6\n",
            "\tgot: 24\n",
            "\ttest: 10\n",
            "\ttry: 17\n",
            "Top 10 tokens in category sportsteam: \n",
            "\tfootball: 7\n",
            "\tgame: 13\n",
            "\tvs: 5\n",
            "\tscores: 3\n",
            "\t@robmoysey: 2\n",
            "\tcheap: 2\n",
            "\tbrooms: 2\n",
            "\t#ryerson: 2\n",
            "\t@ruquidditch: 2\n",
            "\t#rams: 2\n",
            "Top 10 tokens in category tvshow: \n",
            "\t@k8_walsh: 2\n",
            "\tloop: 2\n",
            "\tstupid: 2\n",
            "\tik: 2\n",
            "\tcosta: 2\n",
            "\trican: 2\n",
            "\thttp://ht.ly/2g4ns: 2\n",
            "\t@fuseboxradio: 2\n",
            "\t@planetill: 2\n",
            "\tepisodes: 2\n",
            "Top 10 tokens in category musicartist: \n",
            "\tconcerts: 2\n",
            "\tduring/before: 2\n",
            "\t((: 2\n",
            "\tcosta: 2\n",
            "\trican: 2\n",
            "\thttp://ht.ly/2g4ns: 2\n",
            "\t@fuseboxradio: 2\n",
            "\t@planetill: 2\n",
            "\t@jaredleto: 2\n",
            "\tstage: 2\n"
          ]
        }
      ],
      "source": [
        "TOPN = 10\n",
        "context_sums = {k: (sum(v.values()) * 1.) for k, v in context_words.items()}\n",
        "lobal_sum = len(context_words) * 1.\n",
        "for k, v in context_words.items():\n",
        "  print(\"Top %s tokens in category %s: \" % (TOPN, k))\n",
        "  for t in sorted(v.items(), key=lambda x: ((x[1] / context_sums[k]) * np.log10(global_sum / get_diff_type_occurence(x[0], context_words))), reverse=True)[:TOPN]:\n",
        "    print(\"\\t%s: %s\" % t)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "bX1l3z-pFiN8",
        "outputId": "1393cff3-6866-4e10-8442-9dde1fec1b5f"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-47dc296a-fd7c-4ee3-bcff-3ba79b3ef259\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>token</th>\n",
              "      <th>count</th>\n",
              "      <th>normed_score</th>\n",
              "      <th>normed_global_score</th>\n",
              "      <th>tf_idf</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>163</th>\n",
              "      <td>loop</td>\n",
              "      <td>2</td>\n",
              "      <td>0.003503</td>\n",
              "      <td>1.00000</td>\n",
              "      <td>0.003503</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>160</th>\n",
              "      <td>@k8_walsh</td>\n",
              "      <td>2</td>\n",
              "      <td>0.003503</td>\n",
              "      <td>1.00000</td>\n",
              "      <td>0.003503</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>82</th>\n",
              "      <td>http://ht.ly/2g4ns</td>\n",
              "      <td>2</td>\n",
              "      <td>0.003503</td>\n",
              "      <td>0.69897</td>\n",
              "      <td>0.002448</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>stupid</td>\n",
              "      <td>2</td>\n",
              "      <td>0.003503</td>\n",
              "      <td>0.69897</td>\n",
              "      <td>0.002448</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>252</th>\n",
              "      <td>premiere</td>\n",
              "      <td>2</td>\n",
              "      <td>0.003503</td>\n",
              "      <td>0.69897</td>\n",
              "      <td>0.002448</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>76</th>\n",
              "      <td>costa</td>\n",
              "      <td>2</td>\n",
              "      <td>0.003503</td>\n",
              "      <td>0.69897</td>\n",
              "      <td>0.002448</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>77</th>\n",
              "      <td>rican</td>\n",
              "      <td>2</td>\n",
              "      <td>0.003503</td>\n",
              "      <td>0.69897</td>\n",
              "      <td>0.002448</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>176</th>\n",
              "      <td>episodes</td>\n",
              "      <td>2</td>\n",
              "      <td>0.003503</td>\n",
              "      <td>0.69897</td>\n",
              "      <td>0.002448</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>84</th>\n",
              "      <td>@fuseboxradio</td>\n",
              "      <td>2</td>\n",
              "      <td>0.003503</td>\n",
              "      <td>0.69897</td>\n",
              "      <td>0.002448</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>85</th>\n",
              "      <td>@planetill</td>\n",
              "      <td>2</td>\n",
              "      <td>0.003503</td>\n",
              "      <td>0.69897</td>\n",
              "      <td>0.002448</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-47dc296a-fd7c-4ee3-bcff-3ba79b3ef259')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-840a43e7-7ad9-4248-8800-295aad2ece7e\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-840a43e7-7ad9-4248-8800-295aad2ece7e')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-840a43e7-7ad9-4248-8800-295aad2ece7e button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-47dc296a-fd7c-4ee3-bcff-3ba79b3ef259 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-47dc296a-fd7c-4ee3-bcff-3ba79b3ef259');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "text/plain": [
              "                  token  count  normed_score  normed_global_score    tf_idf\n",
              "163                loop      2      0.003503              1.00000  0.003503\n",
              "160           @k8_walsh      2      0.003503              1.00000  0.003503\n",
              "82   http://ht.ly/2g4ns      2      0.003503              0.69897  0.002448\n",
              "28               stupid      2      0.003503              0.69897  0.002448\n",
              "252            premiere      2      0.003503              0.69897  0.002448\n",
              "76                costa      2      0.003503              0.69897  0.002448\n",
              "77                rican      2      0.003503              0.69897  0.002448\n",
              "176            episodes      2      0.003503              0.69897  0.002448\n",
              "84        @fuseboxradio      2      0.003503              0.69897  0.002448\n",
              "85           @planetill      2      0.003503              0.69897  0.002448"
            ]
          },
          "execution_count": 48,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "phrase_type_key=\"tvshow\"\n",
        "df_t = pd.DataFrame(context_words[phrase_type_key].items(), columns=[\"token\", \"count\"])\n",
        "df_t[\"normed_score\"] = df_t.apply(lambda x: (x[\"count\"] /context_sums[phrase_type_key]), axis=1)\n",
        "df_t[\"normed_global_score\"] = df_t.apply(lambda x: np.log10(global_sum/get_diff_type_occurence(x[0], context_words)), axis=1)\n",
        "df_t[\"tf_idf\"] = df_t[\"normed_score\"]*df_t[\"normed_global_score\"]\n",
        "df_t.sort_values(\"tf_idf\", ascending=False).head(10)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "h-LvZSaILKl6",
        "outputId": "1be00f87-5c31-42f6-8f2c-7d501654b573"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-a404c3ea-8d87-4b3a-aefe-63e0627e3bbf\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>token</th>\n",
              "      <th>count</th>\n",
              "      <th>normed_score</th>\n",
              "      <th>normed_global_score</th>\n",
              "      <th>tf_idf</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>165</th>\n",
              "      <td>...\"</td>\n",
              "      <td>14</td>\n",
              "      <td>0.002070</td>\n",
              "      <td>0.698970</td>\n",
              "      <td>0.001447</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>155</th>\n",
              "      <td>took</td>\n",
              "      <td>17</td>\n",
              "      <td>0.002514</td>\n",
              "      <td>0.522879</td>\n",
              "      <td>0.001314</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>167</th>\n",
              "      <td>part</td>\n",
              "      <td>14</td>\n",
              "      <td>0.002070</td>\n",
              "      <td>0.522879</td>\n",
              "      <td>0.001082</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>321</th>\n",
              "      <td>wake</td>\n",
              "      <td>7</td>\n",
              "      <td>0.001035</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.001035</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>319</th>\n",
              "      <td>concert</td>\n",
              "      <td>10</td>\n",
              "      <td>0.001479</td>\n",
              "      <td>0.698970</td>\n",
              "      <td>0.001034</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>318</th>\n",
              "      <td>trampled</td>\n",
              "      <td>6</td>\n",
              "      <td>0.000887</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000887</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>320</th>\n",
              "      <td>yoiu</td>\n",
              "      <td>6</td>\n",
              "      <td>0.000887</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000887</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>166</th>\n",
              "      <td>got</td>\n",
              "      <td>24</td>\n",
              "      <td>0.003549</td>\n",
              "      <td>0.221849</td>\n",
              "      <td>0.000787</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>382</th>\n",
              "      <td>test</td>\n",
              "      <td>10</td>\n",
              "      <td>0.001479</td>\n",
              "      <td>0.522879</td>\n",
              "      <td>0.000773</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>169</th>\n",
              "      <td>try</td>\n",
              "      <td>17</td>\n",
              "      <td>0.002514</td>\n",
              "      <td>0.301030</td>\n",
              "      <td>0.000757</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a404c3ea-8d87-4b3a-aefe-63e0627e3bbf')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-789313e6-c398-46c3-8610-c1d0c1e2e241\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-789313e6-c398-46c3-8610-c1d0c1e2e241')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-789313e6-c398-46c3-8610-c1d0c1e2e241 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-a404c3ea-8d87-4b3a-aefe-63e0627e3bbf button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-a404c3ea-8d87-4b3a-aefe-63e0627e3bbf');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "text/plain": [
              "        token  count  normed_score  normed_global_score    tf_idf\n",
              "165      ...\"     14      0.002070             0.698970  0.001447\n",
              "155      took     17      0.002514             0.522879  0.001314\n",
              "167      part     14      0.002070             0.522879  0.001082\n",
              "321      wake      7      0.001035             1.000000  0.001035\n",
              "319   concert     10      0.001479             0.698970  0.001034\n",
              "318  trampled      6      0.000887             1.000000  0.000887\n",
              "320      yoiu      6      0.000887             1.000000  0.000887\n",
              "166       got     24      0.003549             0.221849  0.000787\n",
              "382      test     10      0.001479             0.522879  0.000773\n",
              "169       try     17      0.002514             0.301030  0.000757"
            ]
          },
          "execution_count": 49,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "phrase_type_key=\"person\"\n",
        "df_t = pd.DataFrame(context_words[phrase_type_key].items(), columns=[\"token\", \"count\"])\n",
        "df_t[\"normed_score\"] = df_t.apply(lambda x: (x[\"count\"] /context_sums[phrase_type_key]), axis=1)\n",
        "df_t[\"normed_global_score\"] = df_t.apply(lambda x: np.log10(global_sum/get_diff_type_occurence(x[0], context_words)), axis=1)\n",
        "df_t[\"tf_idf\"] = df_t[\"normed_score\"]*df_t[\"normed_global_score\"]\n",
        "df_t.sort_values(\"tf_idf\", ascending=False).head(10)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "ZYjp-WXjLP26",
        "outputId": "cbfd2235-2b62-4b4b-9825-ed4155d7afed"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-7db79aab-ade6-4a28-9828-a15ef8c31e40\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>token</th>\n",
              "      <th>count</th>\n",
              "      <th>normed_score</th>\n",
              "      <th>normed_global_score</th>\n",
              "      <th>tf_idf</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>137</th>\n",
              "      <td>party</td>\n",
              "      <td>10</td>\n",
              "      <td>0.002571</td>\n",
              "      <td>0.397940</td>\n",
              "      <td>0.001023</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>940</th>\n",
              "      <td>follow</td>\n",
              "      <td>6</td>\n",
              "      <td>0.001542</td>\n",
              "      <td>0.522879</td>\n",
              "      <td>0.000806</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>730</th>\n",
              "      <td>http://bit.ly/cds8pd</td>\n",
              "      <td>3</td>\n",
              "      <td>0.000771</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000771</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>341</th>\n",
              "      <td>f</td>\n",
              "      <td>3</td>\n",
              "      <td>0.000771</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000771</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1046</th>\n",
              "      <td>forward</td>\n",
              "      <td>3</td>\n",
              "      <td>0.000771</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000771</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>727</th>\n",
              "      <td>3d</td>\n",
              "      <td>3</td>\n",
              "      <td>0.000771</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000771</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>725</th>\n",
              "      <td>sony/bud</td>\n",
              "      <td>3</td>\n",
              "      <td>0.000771</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000771</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>726</th>\n",
              "      <td>light</td>\n",
              "      <td>3</td>\n",
              "      <td>0.000771</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000771</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37</th>\n",
              "      <td>beautiful</td>\n",
              "      <td>4</td>\n",
              "      <td>0.001028</td>\n",
              "      <td>0.698970</td>\n",
              "      <td>0.000719</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>200</th>\n",
              "      <td>etc</td>\n",
              "      <td>4</td>\n",
              "      <td>0.001028</td>\n",
              "      <td>0.698970</td>\n",
              "      <td>0.000719</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7db79aab-ade6-4a28-9828-a15ef8c31e40')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-01146f14-7a5e-409c-bb59-8e213849a735\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-01146f14-7a5e-409c-bb59-8e213849a735')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-01146f14-7a5e-409c-bb59-8e213849a735 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-7db79aab-ade6-4a28-9828-a15ef8c31e40 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-7db79aab-ade6-4a28-9828-a15ef8c31e40');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "text/plain": [
              "                     token  count  normed_score  normed_global_score    tf_idf\n",
              "137                  party     10      0.002571             0.397940  0.001023\n",
              "940                 follow      6      0.001542             0.522879  0.000806\n",
              "730   http://bit.ly/cds8pd      3      0.000771             1.000000  0.000771\n",
              "341                      f      3      0.000771             1.000000  0.000771\n",
              "1046               forward      3      0.000771             1.000000  0.000771\n",
              "727                     3d      3      0.000771             1.000000  0.000771\n",
              "725               sony/bud      3      0.000771             1.000000  0.000771\n",
              "726                  light      3      0.000771             1.000000  0.000771\n",
              "37               beautiful      4      0.001028             0.698970  0.000719\n",
              "200                    etc      4      0.001028             0.698970  0.000719"
            ]
          },
          "execution_count": 50,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "phrase_type_key=\"geo-loc\"\n",
        "df_t = pd.DataFrame(context_words[phrase_type_key].items(), columns=[\"token\", \"count\"])\n",
        "df_t[\"normed_score\"] = df_t.apply(lambda x: (x[\"count\"] /context_sums[phrase_type_key]), axis=1)\n",
        "df_t[\"normed_global_score\"] = df_t.apply(lambda x: np.log10(global_sum/get_diff_type_occurence(x[0], context_words)), axis=1)\n",
        "df_t[\"tf_idf\"] = df_t[\"normed_score\"]*df_t[\"normed_global_score\"]\n",
        "df_t.sort_values(\"tf_idf\", ascending=False).head(10)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UJRLwv_vWd54"
      },
      "source": [
        "### Questionnaire\n",
        "\n",
        "What other ner data annotation formats are available and how are they different\n",
        "Ans: In addition to CoNLL, the other NER data anootation formats are - JSON, XML, CSV, BIO/BIOES, PASCAL VOC, COCO\n",
        "\n",
        "Why do we need tokenization of the data in our case\n",
        "Tokenization is a fundamental step in NLP that transforms raw text data into a structured and meaningful representation that can be fed into machine learning models for various tasks, such as classification, translation, sentiment analysis, and more.\n",
        "\n",
        "What other models can you use for this task\n",
        "The other models are: Rule-based tokenization, NLTK, SpaCy tokenizer, OpenNLP tokenizer, Transformers library tokenizer, Byte-Pair Encoding (BPE), SentencePiece, Unigram Language Model.\n",
        "\n",
        "Did early stopping have any effect on the training and results.\n",
        "Yes\n",
        "\n",
        "How does the BERT model expect a pair of sentences to be processed?\n",
        "BERT is designed to understand the context of words in a sentence bidirectionally, which means it considers both the left and right context of each word.\n",
        "\n",
        "Why choose Attention based models over Recurrent based ones?\n",
        "Attention-based models have gained popularity due to their ability to handle long-range dependencies, parallelize well, provide interpretability, and facilitate transfer learning.\n",
        "\n",
        "Differentiate BERT and simple transformers\n",
        "BERT is a specific transformer-based language model architecture that is pre-trained on a large corpus of text and fine-tuned on specific tasks. On the other hand, Simple Transformers is a Python library that provides a user-friendly interface to work with transformer-based models like BERT and perform various NLP tasks without dealing with the low-level implementation details of the underlying Transformers library. Simple Transformers makes it easier to use and leverage transformer models for NLP applications.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "qXufw3YfLVL_"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "authorship_tag": "ABX9TyMzBlldPfWXbSYj1wNpvNk2",
      "gpuType": "T4",
      "include_colab_link": true,
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}