Spacy_TAS.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Spacy TAS.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/gustavotr/text-sentiment-datasets/blob/master/Spacy_TAS.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-GdBIaZbu1fj",
        "colab_type": "text"
      },
      "source": [
        "## Faça upload do arquivo \".txt\" que será analisado\n",
        "O texto deve conter uma opinião/review de hotel ou restaurante"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZfwpMvI1vIKL",
        "colab_type": "code",
        "outputId": "017c6f7d-3d05-4815-b6c2-2c804c152189",
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": "OK"
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 76
        }
      },
      "source": [
        "from google.colab import files\n",
        "\n",
        "uploaded = files.upload()\n",
        "\n",
        "#@markdown Arquivo de exemplo pode ser encontrado em https://github.com/gustavotr/text-sentiment-datasets/blob/master/opinioes.txt"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-22355d79-eb8d-4ff9-b798-61ba5576ee68\" name=\"files[]\" multiple disabled />\n",
              "     <output id=\"result-22355d79-eb8d-4ff9-b798-61ba5576ee68\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Saving opinioes.txt to opinioes (7).txt\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f5U3spzh_Dd-",
        "colab_type": "text"
      },
      "source": [
        "## Treinamento para análise de sentimento"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dFn8nskx_Bxx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import json\n",
        "import pandas as pd\n",
        "from sklearn.feature_extraction.text import CountVectorizer,TfidfVectorizer\n",
        "from sklearn.base import TransformerMixin\n",
        "from sklearn.pipeline import Pipeline  "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LM-OZqmbm2Cu",
        "colab_type": "code",
        "outputId": "f85e63a9-0d7b-43b6-eab1-0bba457c36c1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        }
      },
      "source": [
        "#@markdown O dataset utilizado foi criado para esta tarefa retirando do site tripadvisor os reviews de hoteis e restaurantes de curitiba\n",
        "data = pd.read_csv(\"https://raw.githubusercontent.com/gustavotr/text-sentiment-datasets/master/tripadvisor_reviews_curitiba.csv\")\n",
        "data.columns = ['review', 'rating']\n",
        "\n",
        "#@markdown Os ratings do tripadvisor de 1 a 5 são reclassificados em Negativa (-1), Neutra (0) e Positiva (1)\n",
        "di = {\n",
        "    1: -1,\n",
        "    2: -1,\n",
        "    3: 0,\n",
        "    4: 0,\n",
        "    5: 1\n",
        "}\n",
        "data['rating'].replace(di, inplace=True)\n",
        "\n",
        "data.head()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>review</th>\n",
              "      <th>rating</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Se você quer comer uma massa boa e caseira o B...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Uma otima escolha no shopping Barigui Curitiba...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Minha melhor escolha para comemorar aniversári...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>No dia 15/03/19, estivemos no Barolo do Shoppi...</td>\n",
              "      <td>-1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Fui ao Barolo dia 18/4 , comemorar meu anivers...</td>\n",
              "      <td>-1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                              review  rating\n",
              "0  Se você quer comer uma massa boa e caseira o B...       1\n",
              "1  Uma otima escolha no shopping Barigui Curitiba...       1\n",
              "2  Minha melhor escolha para comemorar aniversári...       1\n",
              "3  No dia 15/03/19, estivemos no Barolo do Shoppi...      -1\n",
              "4  Fui ao Barolo dia 18/4 , comemorar meu anivers...      -1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 273
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XQ63n47bC1L6",
        "colab_type": "text"
      },
      "source": [
        "### Separa o texto entre dados de treinamento e dados para validação"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GlidEqhXJDOs",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "train, test = train_test_split(data, test_size=0.33, random_state=42)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IgbmIS4uDFqa",
        "colab_type": "text"
      },
      "source": [
        "### Plota a quantidade de cada categoria que será usada no treinamento"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QHculbnsLhAo",
        "colab_type": "code",
        "outputId": "e1e18cec-d076-41cb-932d-4b78be0223c8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        }
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "fig = plt.figure(figsize=(8,4))\n",
        "sns.barplot(x = train['rating'].unique(), y=train['rating'].value_counts())\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf4AAAD8CAYAAACW0MaaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n4ZckqSP/D+ylbBCRLEeJAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 576x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CJyzr15xJlh0",
        "colab_type": "code",
        "outputId": "7ad75c4f-6c6a-4afd-8fed-509dfa8146bf",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 180
        }
      },
      "source": [
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "from sklearn.base import TransformerMixin\n",
        "from sklearn.pipeline import Pipeline\n",
        "#from sklearn.svm import LinearSVC\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.metrics import accuracy_score\n",
        "!python -m spacy download pt_core_news_sm\n",
        "import string\n",
        "import re\n",
        "import spacy\n",
        "spacy.load('pt_core_news_sm')\n",
        "from spacy.lang.pt import Portuguese\n",
        "parser = Portuguese()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: pt_core_news_sm==2.0.0 from https://github.com/explosion/spacy-models/releases/download/pt_core_news_sm-2.0.0/pt_core_news_sm-2.0.0.tar.gz#egg=pt_core_news_sm==2.0.0 in /usr/local/lib/python3.6/dist-packages (2.0.0)\n",
            "\n",
            "\u001b[93m    Linking successful\u001b[0m\n",
            "    /usr/local/lib/python3.6/dist-packages/pt_core_news_sm -->\n",
            "    /usr/local/lib/python3.6/dist-packages/spacy/data/pt_core_news_sm\n",
            "\n",
            "    You can now load the model via spacy.load('pt_core_news_sm')\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nsWOCGRrCdTY",
        "colab_type": "text"
      },
      "source": [
        "### Cria as funções para tratar o texto"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JXOh9YZLOCUh",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "STOPLIST = spacy.lang.pt.stop_words.STOP_WORDS\n",
        "SYMBOLS = \" \".join(string.punctuation).split(\" \") + [\"-\", \"...\", \"”\", \"”\"]\n",
        "\n",
        "class CleanTextTransformer(TransformerMixin):\n",
        "\n",
        "   def transform(self, X, **transform_params):\n",
        "        return [cleanText(text) for text in X]\n",
        "\n",
        "   def fit(self, X, y=None, **fit_params):\n",
        "        return self\n",
        "\n",
        "def get_params(self, deep=True):\n",
        "        return {}\n",
        "#@markdown Remove caracteres de quebra e retorno de linha e converte texto para letras minúsculas\n",
        "def cleanText(text):\n",
        "    text = text.strip().replace(\"\\n\", \" \").replace(\"\\r\", \" \")\n",
        "    text = text.lower()\n",
        "    return text\n",
        "\n",
        "#@markdown Remove stop words e símbolos do texto e separa as palavras como tokens\n",
        "def tokenizeText(sample):\n",
        "    tokens = parser(sample)\n",
        "    lemmas = []\n",
        "    for tok in tokens:\n",
        "        lemmas.append(tok.lemma_.lower().strip() if tok.lemma_ != \"-PRON-\" else tok.lower_)\n",
        "    tokens = lemmas\n",
        "    tokens = [tok for tok in tokens if tok not in STOPLIST]\n",
        "    tokens = [tok for tok in tokens if tok not in SYMBOLS]\n",
        "    return tokens\n",
        "  "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cKuQJny1GS0d",
        "colab_type": "text"
      },
      "source": [
        "### Realiza o trainamento"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vlxjRniNOjPl",
        "colab_type": "code",
        "outputId": "6df3aeef-9e51-4757-81b4-ded2efffebba",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 537
        }
      },
      "source": [
        "#@markdown Mostra as melhores palavras para determinar a categoria de um texto\n",
        "def printNMostInformative(vectorizer, clf, N):\n",
        "    feature_names = vectorizer.get_feature_names()\n",
        "    coefs_with_fns = sorted(zip(clf.coef_[0], feature_names))\n",
        "    topClass1 = coefs_with_fns[:N]\n",
        "    topClass2 = coefs_with_fns[:-(N + 1):-1]\n",
        "    print(\"Class 1 best: \")\n",
        "    for feat in topClass1:\n",
        "        print(feat)\n",
        "    print(\"Class 2 best: \")\n",
        "    for feat in topClass2:\n",
        "        print(feat)\n",
        "\n",
        "vectorizer = CountVectorizer(tokenizer=tokenizeText, ngram_range=(1,2))\n",
        "#classificador usado\n",
        "clf = LogisticRegression()\n",
        "\n",
        "pipe = Pipeline([('cleanText', CleanTextTransformer()), ('vectorizer', vectorizer), ('clf', clf)])\n",
        "\n",
        "# data\n",
        "train1 = train['review'].tolist()\n",
        "labelsTrain1 = train['rating'].tolist()\n",
        "\n",
        "test1 = test['review'].tolist()\n",
        "labelsTest1 = test['rating'].tolist()\n",
        "# train\n",
        "pipe.fit(train1, labelsTrain1)\n",
        "\n",
        "# test\n",
        "preds = pipe.predict(test1)\n",
        "print(\"accuracy:\", accuracy_score(labelsTest1, preds))\n",
        "print(\"Top 10 features used to predict: \")\n",
        "\n",
        "printNMostInformative(vectorizer, clf, 10)\n",
        "pipe = Pipeline([('cleanText', CleanTextTransformer()), ('vectorizer', vectorizer)])\n",
        "transform = pipe.fit_transform(train1, labelsTrain1)\n",
        "\n",
        "vocab = vectorizer.get_feature_names()\n",
        "for i in range(len(train1)):\n",
        "    s = \"\"\n",
        "    indexIntoVocab = transform.indices[transform.indptr[i]:transform.indptr[i+1]]\n",
        "    numOccurences = transform.data[transform.indptr[i]:transform.indptr[i+1]]\n",
        "    for idx, num in zip(indexIntoVocab, numOccurences):\n",
        "        s += str((vocab[idx], num))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
            "  FutureWarning)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/logistic.py:469: FutureWarning: Default multi_class will be changed to 'auto' in 0.22. Specify the multi_class option to silence this warning.\n",
            "  \"this warning.\", FutureWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "accuracy: 0.751506456241033\n",
            "Top 10 features used to predict: \n",
            "Class 1 best: \n",
            "(-1.7153831761184826, 'excelente')\n",
            "(-1.3292803028532703, 'ótimo')\n",
            "(-1.288424703965177, 'delicioso')\n",
            "(-1.122370684445232, 'agradável')\n",
            "(-0.9870222393321083, 'saboroso')\n",
            "(-0.9771604776894136, 'maravilhoso')\n",
            "(-0.8170212824915254, 'confortável')\n",
            "(-0.7965518008651998, 'perfazer')\n",
            "(-0.6975598407919021, 'gostoso')\n",
            "(-0.6926058034911632, 'limpar')\n",
            "Class 2 best: \n",
            "(2.2216879093849506, 'mau')\n",
            "(1.4663170555299796, 'horrível')\n",
            "(1.3302985345686122, 'ruim')\n",
            "(1.0991063094927136, 'demorar')\n",
            "(1.0106768019415795, 'falto')\n",
            "(0.9808399055461474, 'parecer')\n",
            "(0.9629650581593401, 'atendimento demorar')\n",
            "(0.9441316145328567, 'fraco')\n",
            "(0.9087089506515954, 'decepcionar')\n",
            "(0.8786181041072034, 'absurdo')\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "utwEE3jdGon-",
        "colab_type": "text"
      },
      "source": [
        "### Mostra precisão alcançada com o treinamento"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AU56eT-hPJEn",
        "colab_type": "code",
        "outputId": "c2c3fd3d-b8eb-4a1b-f36e-19c853bcd769",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 196
        }
      },
      "source": [
        "from sklearn import metrics\n",
        "target_names = ratingToText(data['rating'].unique())\n",
        "print(metrics.classification_report(ratingToText(labelsTest1), ratingToText(preds), target_names))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "    Positiva       0.80      0.89      0.84      2198\n",
            "    Negativa       0.81      0.56      0.66       290\n",
            "      Neutra       0.60      0.51      0.55       997\n",
            "\n",
            "    accuracy                           0.75      3485\n",
            "   macro avg       0.74      0.65      0.69      3485\n",
            "weighted avg       0.74      0.75      0.74      3485\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wwjjJJFPGvJK",
        "colab_type": "text"
      },
      "source": [
        "## Realiza a análise do texto enviado"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TNy36J9YvQTM",
        "colab_type": "text"
      },
      "source": [
        "### Extrai texto do arquivo enviado"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "C7WLYpOPvSrZ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "text = \"\"\n",
        "for fn in uploaded.keys():  \n",
        "  text += uploaded[fn].decode(\"utf-8\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TAVDr_uWvU5O",
        "colab_type": "text"
      },
      "source": [
        "### Conta o número de parágrafos"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ggOLGYqbvXNj",
        "colab_type": "code",
        "outputId": "eafd0997-4c41-41d0-b8dd-dab0988428d7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "import re\n",
        "#paragraphs = text.split('\\n')\n",
        "paragraphs = re.split(r'\\r*\\n+\\r*\\n+', text)\n",
        "print(\"Parágrafos: \", len(paragraphs))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Parágrafos:  7\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DCChS1s4vZq2",
        "colab_type": "text"
      },
      "source": [
        "### Conta o número de frases"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VKg6gfJpvdSp",
        "colab_type": "code",
        "outputId": "519ee018-052b-4e30-cd77-d70bb8a11c7f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 71
        }
      },
      "source": [
        "import nltk\n",
        "nltk.download('punkt')\n",
        "\n",
        "sentences = nltk.sent_tokenize(text)\n",
        "print(\"Frases: \", len(sentences))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Package punkt is already up-to-date!\n",
            "Frases:  15\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-Rt0xqCZvj_T",
        "colab_type": "text"
      },
      "source": [
        "### Conta o número de palavras"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8l2WqevTvlvM",
        "colab_type": "code",
        "outputId": "5af4e75d-a64c-491e-ef4d-05b3f66e17e1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "words = nltk.word_tokenize(text)\n",
        "#@markdown Remove símbolos para que não sejam considerados como palavras\n",
        "words = [tok for tok in words if tok not in SYMBOLS]\n",
        "print(\"Palavras: \",len(words))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Palavras:  334\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2BW6sgdwIL-K",
        "colab_type": "text"
      },
      "source": [
        "### Faz a classificação dos sentimentos em cada frase"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "w7wrfVG6TNrS",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#@markdown Renomeia os valores numéricos para melhor visualização\n",
        "ratingsOpts = {\n",
        "    -1: \"Negativa\",\n",
        "    0: \"Neutra\",\n",
        "    1: \"Positiva\"\n",
        "}\n",
        "\n",
        "def ratingToText(ratings):\n",
        "  new_r = []\n",
        "  for r in ratings:\n",
        "    new_r.append(ratingsOpts[r])\n",
        "  return new_r"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OLDjb3qQZNcq",
        "colab_type": "code",
        "outputId": "c5d00594-2fdf-42fc-a980-e7a01e24de9f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        }
      },
      "source": [
        "pipe = Pipeline([('cleanText', CleanTextTransformer()), ('vectorizer', vectorizer), ('clf', clf)])\n",
        "#test2 = ['Quase morri de medo', \"Lugar abafado, sem luz. Tinha um cheiro estranho nos quartos\", \"Não recomendo, horrível\"]\n",
        "preds = pipe.predict(sentences)\n",
        "d = {'review' : sentences, 'sentiment': ratingToText(preds)}\n",
        "predDf = pd.DataFrame(data=d)\n",
        "predDf.head()\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>review</th>\n",
              "      <th>sentiment</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>A casa antiga muito bem cuidada no centro de C...</td>\n",
              "      <td>Neutra</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>As acomodações eram bastante confortáveis, mes...</td>\n",
              "      <td>Neutra</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Banheiros muito limpos\\r\\n\\r\\nA localização do...</td>\n",
              "      <td>Positiva</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Limpeza e atendimento excelentes.</td>\n",
              "      <td>Positiva</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Não tem como não gostar do hostel :)\\r\\n\\r\\nGo...</td>\n",
              "      <td>Positiva</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                              review sentiment\n",
              "0  A casa antiga muito bem cuidada no centro de C...    Neutra\n",
              "1  As acomodações eram bastante confortáveis, mes...    Neutra\n",
              "2  Banheiros muito limpos\\r\\n\\r\\nA localização do...  Positiva\n",
              "3                  Limpeza e atendimento excelentes.  Positiva\n",
              "4  Não tem como não gostar do hostel :)\\r\\n\\r\\nGo...  Positiva"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 285
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XWyemFriG7Cv",
        "colab_type": "text"
      },
      "source": [
        "### Plota o gráfico dos sentimentos no texto"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eqImNTri8mjz",
        "colab_type": "code",
        "outputId": "b6216e6c-731c-41be-c4e3-8e21ff24bb64",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 255
        }
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Data to plot\n",
        "freq = predDf['sentiment'].value_counts()\n",
        "labels = predDf['sentiment'].unique()\n",
        "sizes = [freq[l] for l in labels]\n",
        "color = {'Positiva': 'lightgreen', 'Neutra': 'lightblue', 'Negativa': 'red'}\n",
        "colors = [color[l] for l in labels]\n",
        "\n",
        "# Plot\n",
        "plt.pie(sizes, labels=labels, colors=colors,\n",
        "autopct='%1.1f%%', shadow=True, startangle=140)\n",
        "\n",
        "plt.axis('equal')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWQAAADuCAYAAAAOR30qAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ncobMSDQAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}