ARSA__CODE.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "9c_6IDa9Q1AW"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "review_df=pd.read_csv('C:/Users/vishe/OneDrive/Desktop/Mini_project_final_review/CellPhonesRating.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "tKqQJ6OE66VD"
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "class Lemma4U:\n",
    "    def __init__(self):\n",
    "        self.final_corpus = {}\n",
    "        with open(\"C:/Users/vishe/OneDrive/Desktop/Mini_project_final_review/Custom_Lemmatizer/final.pkl\", 'rb') as f:\n",
    "            self.final_corpus.update(pickle.load(f))\n",
    "\n",
    "    def penn_to_ud(self,tag):\n",
    "        \"\"\"Converting the tags of Penn Tree bank to universal UD tags.\"\"\"\n",
    "\n",
    "        if tag in [\"NN\", \"NNS\"]:\n",
    "            return \"NOUN\"\n",
    "        elif tag in [\"NNP\", \"NNPS\"]:\n",
    "            return \"PROPN\"\n",
    "        elif \"JJ\" in tag or tag == \"AFX\":\n",
    "            return \"ADJ\"\n",
    "        elif tag in [\"#\", \"$\", \"SYM\"]:\n",
    "            return \"SYM\"\n",
    "        elif tag in \"\\\",-LRB--RRB-.:\\'\\'\" or tag == \"HYPH\":\n",
    "            return \"PUNCT\"\n",
    "        elif tag == \"CC\":\n",
    "            return \"CCONJ\"\n",
    "        elif tag == \"CD\":\n",
    "            return \"NUM\"\n",
    "        elif tag in [\"EX\", \"PRP\", \"WP\"]:\n",
    "            return \"PRON\"\n",
    "        elif tag in [\"FW\", \"LS\", \"NIL\"]:\n",
    "            return \"X\"\n",
    "        elif tag in [\"IN\", \"RP\"]:\n",
    "            return \"ADP\"\n",
    "        elif tag in [\"DT\", \"PDT\", \"PRP$\", \"WDT\", \"WP$\"]:\n",
    "            return \"DET\"\n",
    "        elif tag in [\"POS\", \"TO\"]:\n",
    "            return \"PART\"\n",
    "        elif \"RB\" in tag or tag == \"WBR\":\n",
    "            return \"ADV\"\n",
    "        elif tag == \"UH\":\n",
    "            return \"INTJ\"\n",
    "        elif \"VB\" in tag or tag == \"MD\":\n",
    "            return \"VERB\"\n",
    "        else:\n",
    "            return \"X\"\n",
    "\n",
    "    def lemmatize(self, word, pos):\n",
    "\n",
    "        if word in self.final_corpus:\n",
    "            if pos in self.final_corpus[word]:\n",
    "                return self.final_corpus[word][pos]\n",
    "        return word\n",
    "\n",
    "\n",
    "    def plu_to_sing(self,word):\n",
    "        vowels = \"aeiou\"\n",
    "        consonants = \"bcdfghjklmnpqrstwxyz\"\n",
    "        word = str(word).lower()\n",
    "        if len(word) < 2:\n",
    "            return word\n",
    "        if word[-1] == \"s\":\n",
    "            if len(word) > 3:\n",
    "                if word[-3:] == \"ves\":\n",
    "                    return word.replace('ves', 'f')\n",
    "                if word[-3:] == \"ies\":\n",
    "                    return word.replace(\"ies\", 'y')\n",
    "                if word[-2:] == \"es\":\n",
    "                    if word[-3:] == \"ses\" and word[-4] in vowels:\n",
    "                        return word[:-1]\n",
    "                    if word[-4:] == \"zzes\":\n",
    "                        return word.replace('zzes', 'z')\n",
    "                    return word[:-2]\n",
    "                if word[-2:] == \"ys\":\n",
    "                    return word.replace('ys', 'y')\n",
    "                return word[:-1]\n",
    "        return word\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "bC-v_X5U6_nD"
   },
   "outputs": [],
   "source": [
    "obj=Lemma4U()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "id": "VY5jtVEg7B_H",
    "outputId": "9929a6da-5dca-46d7-cd0b-b066b003c6a4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'NOUN'"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj.penn_to_ud(\"NNS\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "id": "6zl4Q6S27GlM",
    "outputId": "5acfb450-57ac-4a25-cb25-75367e562652"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'gun'"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj.plu_to_sing(\"guns\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "id": "FRQ4KsvA7MP3",
    "outputId": "7c5a1487-bda8-47ed-9031-33301e3546c8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'eat'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj.lemmatize(\"ate\",\"VERB\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "X9O3TjtEtlW7"
   },
   "outputs": [],
   "source": [
    "review_df.insert(0,\"SR\",range(0,len(review_df)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 493
    },
    "id": "Hc2k0Nfqwqfj",
    "outputId": "63f634d0-c25d-4c5f-93e7-914d7dc95a9f"
   },
   "outputs": [
    {
     "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>SR</th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>rating</th>\n",
       "      <th>verified</th>\n",
       "      <th>reviewTime</th>\n",
       "      <th>reviewerID</th>\n",
       "      <th>productID</th>\n",
       "      <th>reviewText</th>\n",
       "      <th>summary</th>\n",
       "      <th>unixReviewTime</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>True</td>\n",
       "      <td>08 4, 2014</td>\n",
       "      <td>A24E3SXTC62LJI</td>\n",
       "      <td>7508492919</td>\n",
       "      <td>Looks even better in person. Be careful to not...</td>\n",
       "      <td>Can't stop won't stop looking at it</td>\n",
       "      <td>1407110400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>True</td>\n",
       "      <td>02 12, 2014</td>\n",
       "      <td>A269FLZCB4GIPV</td>\n",
       "      <td>7508492919</td>\n",
       "      <td>When you don't want to spend a whole lot of ca...</td>\n",
       "      <td>1</td>\n",
       "      <td>1392163200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3.0</td>\n",
       "      <td>True</td>\n",
       "      <td>02 8, 2014</td>\n",
       "      <td>AB6CHQWHZW4TV</td>\n",
       "      <td>7508492919</td>\n",
       "      <td>so the case came on time, i love the design. I...</td>\n",
       "      <td>Its okay</td>\n",
       "      <td>1391817600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>True</td>\n",
       "      <td>02 4, 2014</td>\n",
       "      <td>A1M117A53LEI8</td>\n",
       "      <td>7508492919</td>\n",
       "      <td>DON'T CARE FOR IT.  GAVE IT AS A GIFT AND THEY...</td>\n",
       "      <td>CASE</td>\n",
       "      <td>1391472000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4.0</td>\n",
       "      <td>True</td>\n",
       "      <td>02 3, 2014</td>\n",
       "      <td>A272DUT8M88ZS8</td>\n",
       "      <td>7508492919</td>\n",
       "      <td>I liked it because it was cute, but the studs ...</td>\n",
       "      <td>Cute!</td>\n",
       "      <td>1391385600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>938256</th>\n",
       "      <td>938256</td>\n",
       "      <td>1128431</td>\n",
       "      <td>3.0</td>\n",
       "      <td>True</td>\n",
       "      <td>01 2, 2017</td>\n",
       "      <td>A67N8PO0ZMYC5</td>\n",
       "      <td>B01HJC7N4C</td>\n",
       "      <td>my sons likes it. For the price it isn't too bad</td>\n",
       "      <td>Three Stars</td>\n",
       "      <td>1483315200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>938257</th>\n",
       "      <td>938257</td>\n",
       "      <td>1128432</td>\n",
       "      <td>4.0</td>\n",
       "      <td>True</td>\n",
       "      <td>12 22, 2016</td>\n",
       "      <td>A1QWMCG1FNEP3A</td>\n",
       "      <td>B01HJC7N4C</td>\n",
       "      <td>Good for viewing. But doesn't have a button or...</td>\n",
       "      <td>Good</td>\n",
       "      <td>1482364800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>938258</th>\n",
       "      <td>938258</td>\n",
       "      <td>1128433</td>\n",
       "      <td>5.0</td>\n",
       "      <td>False</td>\n",
       "      <td>07 15, 2016</td>\n",
       "      <td>A3FOBEJ9UVUTR3</td>\n",
       "      <td>B01HJC7N4C</td>\n",
       "      <td>I was given the Rockrok 3D VR Glasses Headset ...</td>\n",
       "      <td>THE FUTURE IS NOW!!!!!!!</td>\n",
       "      <td>1468540800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>938259</th>\n",
       "      <td>938259</td>\n",
       "      <td>1128434</td>\n",
       "      <td>5.0</td>\n",
       "      <td>False</td>\n",
       "      <td>07 14, 2016</td>\n",
       "      <td>AMUEAMKB4E33M</td>\n",
       "      <td>B01HJC7N4C</td>\n",
       "      <td>Super Fun! The RockRoc 3d vr headset is waaaay...</td>\n",
       "      <td>Get more out of your smartphone .......</td>\n",
       "      <td>1468454400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>938260</th>\n",
       "      <td>938260</td>\n",
       "      <td>1128435</td>\n",
       "      <td>5.0</td>\n",
       "      <td>False</td>\n",
       "      <td>07 13, 2016</td>\n",
       "      <td>A2EV91MMOJ3IL4</td>\n",
       "      <td>B01HJC7N4C</td>\n",
       "      <td>Love it!\\n\\nI've had other VR glasses which al...</td>\n",
       "      <td>Join the VR fun train!</td>\n",
       "      <td>1468368000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>938261 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            SR  Unnamed: 0  rating  verified   reviewTime      reviewerID  \\\n",
       "0            0           0     5.0      True   08 4, 2014  A24E3SXTC62LJI   \n",
       "1            1           1     5.0      True  02 12, 2014  A269FLZCB4GIPV   \n",
       "2            2           2     3.0      True   02 8, 2014   AB6CHQWHZW4TV   \n",
       "3            3           3     2.0      True   02 4, 2014   A1M117A53LEI8   \n",
       "4            4           4     4.0      True   02 3, 2014  A272DUT8M88ZS8   \n",
       "...        ...         ...     ...       ...          ...             ...   \n",
       "938256  938256     1128431     3.0      True   01 2, 2017   A67N8PO0ZMYC5   \n",
       "938257  938257     1128432     4.0      True  12 22, 2016  A1QWMCG1FNEP3A   \n",
       "938258  938258     1128433     5.0     False  07 15, 2016  A3FOBEJ9UVUTR3   \n",
       "938259  938259     1128434     5.0     False  07 14, 2016   AMUEAMKB4E33M   \n",
       "938260  938260     1128435     5.0     False  07 13, 2016  A2EV91MMOJ3IL4   \n",
       "\n",
       "         productID                                         reviewText  \\\n",
       "0       7508492919  Looks even better in person. Be careful to not...   \n",
       "1       7508492919  When you don't want to spend a whole lot of ca...   \n",
       "2       7508492919  so the case came on time, i love the design. I...   \n",
       "3       7508492919  DON'T CARE FOR IT.  GAVE IT AS A GIFT AND THEY...   \n",
       "4       7508492919  I liked it because it was cute, but the studs ...   \n",
       "...            ...                                                ...   \n",
       "938256  B01HJC7N4C   my sons likes it. For the price it isn't too bad   \n",
       "938257  B01HJC7N4C  Good for viewing. But doesn't have a button or...   \n",
       "938258  B01HJC7N4C  I was given the Rockrok 3D VR Glasses Headset ...   \n",
       "938259  B01HJC7N4C  Super Fun! The RockRoc 3d vr headset is waaaay...   \n",
       "938260  B01HJC7N4C  Love it!\\n\\nI've had other VR glasses which al...   \n",
       "\n",
       "                                        summary  unixReviewTime  \n",
       "0           Can't stop won't stop looking at it      1407110400  \n",
       "1                                             1      1392163200  \n",
       "2                                      Its okay      1391817600  \n",
       "3                                          CASE      1391472000  \n",
       "4                                         Cute!      1391385600  \n",
       "...                                         ...             ...  \n",
       "938256                              Three Stars      1483315200  \n",
       "938257                                     Good      1482364800  \n",
       "938258                 THE FUTURE IS NOW!!!!!!!      1468540800  \n",
       "938259  Get more out of your smartphone .......      1468454400  \n",
       "938260                   Join the VR fun train!      1468368000  \n",
       "\n",
       "[938261 rows x 10 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "review_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "id": "NAAQWewY3u7B"
   },
   "outputs": [],
   "source": [
    "li=list()\n",
    "for i in review_df[\"SR\"]:\n",
    "    sentance=review_df[\"reviewText\"][i]\n",
    "    if type(sentance) is str:\n",
    "        words=sentance.split()\n",
    "        if len(words) > 5:\n",
    "            if len(words) < 100:\n",
    "                if review_df[\"rating\"][i]>3:\n",
    "                    li.append((words,\"pos\"))\n",
    "                else:\n",
    "                    li.append((words,\"neg\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ep61SfX738y1",
    "outputId": "724950d8-0e9e-44ab-d980-df6265af30c4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(705105,\n",
       " ['When',\n",
       "  'you',\n",
       "  \"don't\",\n",
       "  'want',\n",
       "  'to',\n",
       "  'spend',\n",
       "  'a',\n",
       "  'whole',\n",
       "  'lot',\n",
       "  'of',\n",
       "  'cash',\n",
       "  'but',\n",
       "  'want',\n",
       "  'a',\n",
       "  'great',\n",
       "  'deal...this',\n",
       "  'is',\n",
       "  'the',\n",
       "  'shop',\n",
       "  'to',\n",
       "  'buy',\n",
       "  'from!'])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(li),li[1][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "zHVac7NLxzdq"
   },
   "outputs": [],
   "source": [
    "import random\n",
    "random.shuffle(li)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "H9OlzWQ1tAxv"
   },
   "outputs": [],
   "source": [
    "len(li),li[1][0]\n",
    "pd.options.display.max_rows = 10000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "86lxjzSCo3vv",
    "outputId": "2566f283-7927-41d0-da5a-c1ab138750b7"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(705105,\n",
       " (['I',\n",
       "   'used',\n",
       "   'this',\n",
       "   'to',\n",
       "   'replace',\n",
       "   'the',\n",
       "   'power',\n",
       "   'supply',\n",
       "   'of',\n",
       "   'my',\n",
       "   'new',\n",
       "   'NEST',\n",
       "   'Outdoor',\n",
       "   'Camera.',\n",
       "   'I',\n",
       "   'did',\n",
       "   'not',\n",
       "   'need',\n",
       "   'all',\n",
       "   'the',\n",
       "   'extra',\n",
       "   'wiring',\n",
       "   'and',\n",
       "   'after',\n",
       "   'searching',\n",
       "   'for',\n",
       "   'a',\n",
       "   'while',\n",
       "   'this',\n",
       "   'product',\n",
       "   'was',\n",
       "   'the',\n",
       "   'only',\n",
       "   'one',\n",
       "   'that',\n",
       "   'meet',\n",
       "   'the',\n",
       "   'NEST',\n",
       "   'power',\n",
       "   'supply',\n",
       "   'specs',\n",
       "   'allowing',\n",
       "   'me',\n",
       "   'to',\n",
       "   'replace',\n",
       "   'the',\n",
       "   'extra',\n",
       "   '10',\n",
       "   'feet',\n",
       "   'of',\n",
       "   'cord',\n",
       "   'that',\n",
       "   'I',\n",
       "   'did',\n",
       "   'not',\n",
       "   'need.',\n",
       "   'This',\n",
       "   'helped',\n",
       "   'make',\n",
       "   'for',\n",
       "   'a',\n",
       "   'much',\n",
       "   'cleaner',\n",
       "   'install!.'],\n",
       "  'pos'),\n",
       " 705105,\n",
       " 64)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y=list()\n",
    "for x in li:\n",
    "    y.append(len(x[0]))\n",
    "len(li),li[1],len(y),y[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "5NogrYL0xqV1"
   },
   "outputs": [],
   "source": [
    "from nltk.stem import WordNetLemmatizer\n",
    "lemmatizer= WordNetLemmatizer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "b0URR2KjyGqt"
   },
   "outputs": [],
   "source": [
    "from nltk.corpus import wordnet\n",
    "def get_simple_pos(tag):\n",
    "    if tag.startswith(\"J\"):\n",
    "        return wordnet.ADJ\n",
    "    elif tag.startswith(\"V\"):\n",
    "        return wordnet.VERB\n",
    "    elif tag.startswith(\"N\"):\n",
    "        return wordnet.NOUN\n",
    "    elif tag.startswith(\"R\"):\n",
    "        return wordnet.ADV\n",
    "    else:\n",
    "        return wordnet.NOUN"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1wjawKXcydUK"
   },
   "source": [
    "import nltk\n",
    "\n",
    "nltk.download('averaged_perceptron_tagger')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6mmdRVkh0LH-",
    "outputId": "5ac3edd9-8870-4628-f4ac-767415f04b20"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('better', 'RBR'), ('bitter', 'NN'), ('butter', 'NN')]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from nltk import pos_tag\n",
    "w=['better','bitter','butter']\n",
    "pos_tag(w)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "q7V53vtG0EMa"
   },
   "outputs": [],
   "source": [
    "from nltk.corpus import stopwords\n",
    "stop=stopwords.words(\"english\")\n",
    "import string \n",
    "punctuations = list(string.punctuation)\n",
    "stop=stop + punctuations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "6CDkVlqz0TlJ"
   },
   "outputs": [],
   "source": [
    "def clean_review(words):\n",
    "    output_words = []\n",
    "    for w in words:\n",
    "        if w.lower() not in stop :\n",
    "            pos = pos_tag(w)\n",
    "            clean_words=lemmatizer.lemmatize(w,pos= get_simple_pos(pos[0][1]))#[0]will give tuple #[1] will be noun\n",
    "            output_words.append(clean_words.lower())\n",
    "    return output_words"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gDKDhBS40_LG"
   },
   "source": [
    "import nltk\n",
    "\n",
    "nltk.download('wordnet')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "FzhvtgSb-T9U",
    "outputId": "f02db5a0-3fa6-489c-b91d-2c362c82ee76"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(['I',\n",
       "  'purchased',\n",
       "  'two',\n",
       "  'of',\n",
       "  'these.',\n",
       "  'The',\n",
       "  'first',\n",
       "  'one',\n",
       "  \"wouldn't\",\n",
       "  'power',\n",
       "  'on',\n",
       "  'even',\n",
       "  'after',\n",
       "  'charging',\n",
       "  'it',\n",
       "  'for',\n",
       "  '7',\n",
       "  'hours.',\n",
       "  'The',\n",
       "  'second',\n",
       "  'one',\n",
       "  'overheated',\n",
       "  'when',\n",
       "  'it',\n",
       "  'reached',\n",
       "  'the',\n",
       "  '4th',\n",
       "  'charge',\n",
       "  'indicator.',\n",
       "  'The',\n",
       "  'plug',\n",
       "  'was',\n",
       "  'very',\n",
       "  'hot',\n",
       "  'and',\n",
       "  'the',\n",
       "  'power',\n",
       "  'bank',\n",
       "  'port',\n",
       "  'had',\n",
       "  'smoke',\n",
       "  'coming',\n",
       "  'from',\n",
       "  'it.',\n",
       "  'I',\n",
       "  'only',\n",
       "  'noticed',\n",
       "  'it',\n",
       "  'because',\n",
       "  'I',\n",
       "  'smelled',\n",
       "  'melted',\n",
       "  'plastic.',\n",
       "  'This',\n",
       "  'product',\n",
       "  'is',\n",
       "  'junk.',\n",
       "  'I',\n",
       "  'thought',\n",
       "  'it',\n",
       "  'was',\n",
       "  'good',\n",
       "  'because',\n",
       "  'it',\n",
       "  'had',\n",
       "  'the',\n",
       "  'features',\n",
       "  'I',\n",
       "  'was',\n",
       "  'looking',\n",
       "  'for',\n",
       "  'at',\n",
       "  'a',\n",
       "  'very',\n",
       "  'low',\n",
       "  'price.',\n",
       "  'The',\n",
       "  'product',\n",
       "  \"doesn't\",\n",
       "  'even',\n",
       "  'have',\n",
       "  'a',\n",
       "  'name',\n",
       "  'on',\n",
       "  'the',\n",
       "  'power',\n",
       "  'bank.'],\n",
       " 'neg')"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x=li[:40000]\n",
    "x[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "id": "Nfj77ZUt0ps7"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "40000"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "review_clean=[(clean_review(review),category) for review,category in x]\n",
    "#if take too long for now we will use x which is x=li[:5000]\n",
    "len(review_clean)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "c-QHGN2H_Lw_"
   },
   "source": [
    "  We faced a serious challenge in this part as our pc's compution power was less we cant simiply go on reading 7000k review so to save our progress we came up with a idea to save our progress using **CSV** libary.\n",
    "  \n",
    "  \n",
    "  w- stands for **writing **\n",
    "\n",
    "  saved as **['clean_review','emotion']**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "uj2vAkK55DRq"
   },
   "outputs": [],
   "source": [
    "import csv\n",
    "data=review_clean\n",
    "with open('review_clean_countvectorize.csv','w') as out:\n",
    "    csv_out=csv.writer(out)\n",
    "    csv_out.writerow(['clean_review','emotion'])\n",
    "    for row in data:\n",
    "        csv_out.writerow(row)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "categories=[category for document,category in review_clean]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "text_reviews=[\" \".join(review) for review ,category in review_clean]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('great case really fit well.', 'pos')"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text_reviews[1],categories[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train,x_test,y_train,y_test = train_test_split(text_reviews,categories)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[0, 0, 0, ..., 0, 0, 0],\n",
       "        [0, 0, 0, ..., 0, 0, 0],\n",
       "        [0, 0, 0, ..., 0, 0, 0],\n",
       "        ...,\n",
       "        [0, 0, 0, ..., 0, 0, 0],\n",
       "        [0, 0, 0, ..., 0, 0, 0],\n",
       "        [0, 0, 0, ..., 0, 0, 0]], dtype=int64)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "count_vec= CountVectorizer(max_features=3000)\n",
    "x_train_features= count_vec.fit_transform(x_train)\n",
    "x_train_features.todense()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('count_vec_(unigram)_model','wb') as f:\n",
    "    pickle.dump(count_vec,f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[0, 0, 0, ..., 0, 0, 0],\n",
       "        [0, 0, 0, ..., 0, 0, 0],\n",
       "        [0, 0, 0, ..., 0, 0, 0],\n",
       "        ...,\n",
       "        [0, 0, 0, ..., 0, 0, 0],\n",
       "        [0, 0, 0, ..., 0, 0, 0],\n",
       "        [0, 0, 0, ..., 0, 0, 0]], dtype=int64)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "count_vec_bigram= CountVectorizer(max_features=6000, ngram_range=(1,2))\n",
    "x_train_features_bigram=count_vec_bigram.fit_transform(x_train)\n",
    "x_train_features_bigram.todense()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('count_vec_(bigram)_model','wb') as f:\n",
    "    pickle.dump(count_vec_bigram,f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "x=count_vec.get_feature_names()\n",
    "with open('sample words unigram.txt', 'w') as filehandle:\n",
    "    for listitem in x:\n",
    "        filehandle.write('%s\\n' % listitem)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "y=count_vec_bigram.get_feature_names()\n",
    "with open('sample words bigram.txt', 'w') as filehandle:\n",
    "    for listitem in y:\n",
    "        filehandle.write('%s\\n' % listitem)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_test_features=count_vec.transform(x_test)\n",
    "x_test_features_bigram=count_vec_bigram.transform(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<10000x3000 sparse matrix of type '<class 'numpy.int64'>'\n",
       " \twith 148698 stored elements in Compressed Sparse Row format>,\n",
       " <10000x6000 sparse matrix of type '<class 'numpy.int64'>'\n",
       " \twith 188879 stored elements in Compressed Sparse Row format>)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_test_features,x_test_features_bigram"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.svm import SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC()"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "C = SVC()\n",
    "svc.fit(x_train_features,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8549"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc.score(x_test_features,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'svc' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-53-1547f8993b68>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0my_pred_uni\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0msvc\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx_test_features\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mcm\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mconfusion_matrix\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0my_test\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0my_pred_uni\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[0mcm\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'svc' is not defined"
     ]
    }
   ],
   "source": [
    "y_pred_uni=svc.predict(x_test_features)\n",
    "cm=confusion_matrix(y_test,y_pred_uni)\n",
    "cm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         neg       0.83      0.51      0.63      2446\n",
      "         pos       0.86      0.97      0.91      7554\n",
      "\n",
      "    accuracy                           0.85     10000\n",
      "   macro avg       0.84      0.74      0.77     10000\n",
      "weighted avg       0.85      0.85      0.84     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,y_pred_uni))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(69.0, 0.5, 'Truth')"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# import matplotlib.pyplot as plt\n",
    "# import seaborn as sn\n",
    "# plt.figure(figsize=(10,7))\n",
    "# sn.heatmap(cm,annot=True)\n",
    "# plt.xlabel(\"Predicted\")\n",
    "# plt.ylabel(\"Truth\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict_uni_trans(Review):\n",
    "    words=Review.split()\n",
    "    clean_words=clean_review(words)\n",
    "    x_paragraph=[\" \".join(clean_words)]\n",
    "    #print(x_paragraph)\n",
    "    x=count_vec.transform(x_paragraph)\n",
    "    #print(x)\n",
    "    return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Review: very bad had to return\n",
      "['neg']\n"
     ]
    }
   ],
   "source": [
    "Review = input(\"Review: \")\n",
    "#print(predict(Review))\n",
    "print(svc.predict(predict_uni_trans(Review)))\n",
    "#I am really happy with the awesome product quality it works really well"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC()"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_bigram= SVC()\n",
    "svc_bigram.fit(x_train_features_bigram,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8593"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_bigram.score(x_test_features_bigram,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'svc_bigram' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-51-ec4af750c525>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0my_pred_bi\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0msvc_bigram\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx_test_features_bigram\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mcm_bi\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mconfusion_matrix\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0my_test\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0my_pred_bi\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[0mcm_bi\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'svc_bigram' is not defined"
     ]
    }
   ],
   "source": [
    "y_pred_bi=svc_bigram.predict(x_test_features_bigram)\n",
    "cm_bi=confusion_matrix(y_test,y_pred_bi)\n",
    "cm_bi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         neg       0.83      0.53      0.65      2446\n",
      "         pos       0.86      0.97      0.91      7554\n",
      "\n",
      "    accuracy                           0.86     10000\n",
      "   macro avg       0.85      0.75      0.78     10000\n",
      "weighted avg       0.86      0.86      0.85     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,y_pred_bi))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(33.0, 0.5, 'Truth')"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.heatmap(cm_bi,annot=True)\n",
    "plt.xlabel(\"Predicted\")\n",
    "plt.ylabel(\"Truth\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict_bi_trans(Review):\n",
    "    words=Review.split()\n",
    "    clean_words=clean_review(words)\n",
    "    x_paragraph=[\" \".join(clean_words)]\n",
    "    #print(x_paragraph)\n",
    "    x=count_vec_bigram.transform(x_paragraph)\n",
    "    #print(x)\n",
    "    return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Review: very bad had to return\n",
      "['neg']\n"
     ]
    }
   ],
   "source": [
    "Review = input(\"Review: \")\n",
    "#print(predict(Review))\n",
    "print(svc_bigram.predict(predict_bi_trans(Review)))\n",
    "#I am really happy with the awesome product quality it works really well"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('SVC_countvectorize_85.93_(bigram)_model','wb') as f:\n",
    "    pickle.dump(svc_bigram,f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('SVC_countvectorize_85.49_(unigram)_model','wb') as f:\n",
    "    pickle.dump(svc,f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import tree\n",
    "Dt_clf_uni= tree.DecisionTreeClassifier()\n",
    "Dt_clf_bi=tree.DecisionTreeClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier()"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Dt_clf_uni.fit(x_train_features,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7809"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Dt_clf_uni.score(x_test_features,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1353, 1107],\n",
       "       [1084, 6456]], dtype=int64)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred_uni=Dt_clf_uni.predict(x_test_features)\n",
    "cm=confusion_matrix(y_test,y_pred_uni)\n",
    "cm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         neg       0.56      0.55      0.55      2460\n",
      "         pos       0.85      0.86      0.85      7540\n",
      "\n",
      "    accuracy                           0.78     10000\n",
      "   macro avg       0.70      0.70      0.70     10000\n",
      "weighted avg       0.78      0.78      0.78     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,y_pred_uni))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Review = input(\"Review: \")\n",
    "#print(predict(Review))\n",
    "print(Dt_clf_uni.predict(predict_uni_trans(Review)))\n",
    "#I am really happy with the awesome product quality it works really well"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier()"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Dt_clf_bi.fit(x_train_features_bigram,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7851"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Dt_clf_bi.score(x_test_features_bigram,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1303, 1143],\n",
       "       [1006, 6548]], dtype=int64)"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred_bi=Dt_clf_bi.predict(x_test_features_bigram)\n",
    "cm_bi=confusion_matrix(y_test,y_pred_bi)\n",
    "cm_bi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         neg       0.56      0.53      0.55      2446\n",
      "         pos       0.85      0.87      0.86      7554\n",
      "\n",
      "    accuracy                           0.79     10000\n",
      "   macro avg       0.71      0.70      0.70     10000\n",
      "weighted avg       0.78      0.79      0.78     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,y_pred_bi))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Review: I am really happy with the awesome product quality it works really well\n",
      "['pos']\n"
     ]
    }
   ],
   "source": [
    "Review = input(\"Review: \")\n",
    "#print(predict(Review))\n",
    "print(Dt_clf_bi.predict(predict_bi_trans(Review)))\n",
    "#I am really happy with the awesome product quality it works really well"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('DT_countvectorize_78.24_(unigram)_model','wb') as f:\n",
    "    pickle.dump(Dt_clf_uni,f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('DT_countvectorize_78.51_(bigram)_model','wb') as f:\n",
    "    pickle.dump(Dt_clf_bi,f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['00',\n",
       " '000',\n",
       " '10',\n",
       " '100',\n",
       " '10ft',\n",
       " '11',\n",
       " '12',\n",
       " '12v',\n",
       " '13',\n",
       " '14',\n",
       " '15',\n",
       " '1520',\n",
       " '16',\n",
       " '17',\n",
       " '18',\n",
       " '19',\n",
       " '1a',\n",
       " '1st',\n",
       " '20',\n",
       " '200',\n",
       " '2013',\n",
       " '2014',\n",
       " '2015',\n",
       " '2016',\n",
       " '2017',\n",
       " '21',\n",
       " '24',\n",
       " '25',\n",
       " '2nd',\n",
       " '30',\n",
       " '32',\n",
       " '35',\n",
       " '360',\n",
       " '3d',\n",
       " '3ft',\n",
       " '3g',\n",
       " '3gs',\n",
       " '3m',\n",
       " '3rd',\n",
       " '40',\n",
       " '42mm',\n",
       " '45',\n",
       " '4g',\n",
       " '4s',\n",
       " '4th',\n",
       " '50',\n",
       " '520',\n",
       " '5c',\n",
       " '5mm',\n",
       " '5s',\n",
       " '5th',\n",
       " '5v',\n",
       " '5x',\n",
       " '60',\n",
       " '635',\n",
       " '6ft',\n",
       " '6p',\n",
       " '6plus',\n",
       " '6s',\n",
       " '70',\n",
       " '80',\n",
       " '90',\n",
       " '95',\n",
       " '99',\n",
       " 'ability',\n",
       " 'able',\n",
       " 'about',\n",
       " 'absolute',\n",
       " 'absolutely',\n",
       " 'absorption',\n",
       " 'abuse',\n",
       " 'ac',\n",
       " 'accent',\n",
       " 'accept',\n",
       " 'acceptable',\n",
       " 'access',\n",
       " 'accessible',\n",
       " 'accessories',\n",
       " 'accessory',\n",
       " 'accident',\n",
       " 'accidental',\n",
       " 'accidentally',\n",
       " 'accidently',\n",
       " 'accommodate',\n",
       " 'according',\n",
       " 'account',\n",
       " 'accurate',\n",
       " 'across',\n",
       " 'act',\n",
       " 'action',\n",
       " 'activate',\n",
       " 'activated',\n",
       " 'activation',\n",
       " 'active',\n",
       " 'activity',\n",
       " 'actual',\n",
       " 'actually',\n",
       " 'ad',\n",
       " 'adapter',\n",
       " 'adapters',\n",
       " 'adaptor',\n",
       " 'add',\n",
       " 'added',\n",
       " 'adding',\n",
       " 'addition',\n",
       " 'additional',\n",
       " 'adds',\n",
       " 'adequate',\n",
       " 'adhere',\n",
       " 'adhesive',\n",
       " 'adjust',\n",
       " 'adjustable',\n",
       " 'adjusted',\n",
       " 'adjusting',\n",
       " 'adjustment',\n",
       " 'admit',\n",
       " 'adorable',\n",
       " 'advantage',\n",
       " 'advertised',\n",
       " 'advertized',\n",
       " 'advice',\n",
       " 'advise',\n",
       " 'aesthetic',\n",
       " 'affect',\n",
       " 'affected',\n",
       " 'affordable',\n",
       " 'afraid',\n",
       " 'after',\n",
       " 'aftermarket',\n",
       " 'again',\n",
       " 'age',\n",
       " 'ago',\n",
       " 'agree',\n",
       " 'ahead',\n",
       " 'air',\n",
       " 'alarm',\n",
       " 'alcohol',\n",
       " 'align',\n",
       " 'aligned',\n",
       " 'alignment',\n",
       " 'all',\n",
       " 'allow',\n",
       " 'allowed',\n",
       " 'allowing',\n",
       " 'allows',\n",
       " 'almost',\n",
       " 'alone',\n",
       " 'along',\n",
       " 'alot',\n",
       " 'already',\n",
       " 'alright',\n",
       " 'also',\n",
       " 'alternative',\n",
       " 'although',\n",
       " 'altogether',\n",
       " 'aluminum',\n",
       " 'always',\n",
       " 'am',\n",
       " 'amazed',\n",
       " 'amazing',\n",
       " 'amazingly',\n",
       " 'amazon',\n",
       " 'amount',\n",
       " 'amp',\n",
       " 'amperage',\n",
       " 'ample',\n",
       " 'and',\n",
       " 'android',\n",
       " 'angle',\n",
       " 'angles',\n",
       " 'anker',\n",
       " 'annoying',\n",
       " 'another',\n",
       " 'answer',\n",
       " 'antenna',\n",
       " 'anti',\n",
       " 'any',\n",
       " 'anybody',\n",
       " 'anymore',\n",
       " 'anyone',\n",
       " 'anything',\n",
       " 'anytime',\n",
       " 'anyway',\n",
       " 'anyways',\n",
       " 'anywhere',\n",
       " 'apart',\n",
       " 'app',\n",
       " 'apparently',\n",
       " 'appear',\n",
       " 'appearance',\n",
       " 'appeared',\n",
       " 'appears',\n",
       " 'apple',\n",
       " 'application',\n",
       " 'applied',\n",
       " 'apply',\n",
       " 'applying',\n",
       " 'appreciate',\n",
       " 'appropriate',\n",
       " 'apps',\n",
       " 'are',\n",
       " 'area',\n",
       " 'arm',\n",
       " 'armband',\n",
       " 'armor',\n",
       " 'armorsuit',\n",
       " 'arms',\n",
       " 'around',\n",
       " 'arrival',\n",
       " 'arrive',\n",
       " 'arrived',\n",
       " 'as',\n",
       " 'asap',\n",
       " 'aside',\n",
       " 'ask',\n",
       " 'asked',\n",
       " 'asking',\n",
       " 'aspect',\n",
       " 'assemble',\n",
       " 'assembly',\n",
       " 'assuming',\n",
       " 'asus',\n",
       " 'at',\n",
       " 'atrix',\n",
       " 'att',\n",
       " 'attach',\n",
       " 'attache',\n",
       " 'attached',\n",
       " 'attaching',\n",
       " 'attachment',\n",
       " 'attempt',\n",
       " 'attention',\n",
       " 'attract',\n",
       " 'attractive',\n",
       " 'attracts',\n",
       " 'audio',\n",
       " 'aukey',\n",
       " 'authentic',\n",
       " 'auto',\n",
       " 'automatically',\n",
       " 'aux',\n",
       " 'available',\n",
       " 'average',\n",
       " 'avoid',\n",
       " 'aware',\n",
       " 'away',\n",
       " 'awesome',\n",
       " 'awful',\n",
       " 'awhile',\n",
       " 'awkward',\n",
       " 'awsome',\n",
       " 'baby',\n",
       " 'back',\n",
       " 'background',\n",
       " 'backpack',\n",
       " 'backside',\n",
       " 'backup',\n",
       " 'bad',\n",
       " 'badly',\n",
       " 'bag',\n",
       " 'balance',\n",
       " 'ball',\n",
       " 'ballistic',\n",
       " 'band',\n",
       " 'bands',\n",
       " 'bang',\n",
       " 'bank',\n",
       " 'bar',\n",
       " 'barely',\n",
       " 'bargain',\n",
       " 'base',\n",
       " 'based',\n",
       " 'basic',\n",
       " 'basically',\n",
       " 'basics',\n",
       " 'basis',\n",
       " 'bass',\n",
       " 'bathroom',\n",
       " 'batter',\n",
       " 'batteries',\n",
       " 'battery',\n",
       " 'bb',\n",
       " 'bc',\n",
       " 'be',\n",
       " 'beach',\n",
       " 'bear',\n",
       " 'beat',\n",
       " 'beating',\n",
       " 'beautiful',\n",
       " 'beautifully',\n",
       " 'beauty',\n",
       " 'became',\n",
       " 'become',\n",
       " 'becomes',\n",
       " 'becoming',\n",
       " 'bed',\n",
       " 'bedside',\n",
       " 'before',\n",
       " 'began',\n",
       " 'begin',\n",
       " 'beginning',\n",
       " 'behind',\n",
       " 'believe',\n",
       " 'belkin',\n",
       " 'belt',\n",
       " 'bend',\n",
       " 'bending',\n",
       " 'benefit',\n",
       " 'bent',\n",
       " 'besides',\n",
       " 'best',\n",
       " 'better',\n",
       " 'between',\n",
       " 'beveled',\n",
       " 'beware',\n",
       " 'beyond',\n",
       " 'bezel',\n",
       " 'big',\n",
       " 'bigger',\n",
       " 'biggest',\n",
       " 'bike',\n",
       " 'bill',\n",
       " 'birthday',\n",
       " 'bit',\n",
       " 'black',\n",
       " 'blackberry',\n",
       " 'blason',\n",
       " 'blaze',\n",
       " 'bling',\n",
       " 'block',\n",
       " 'blocked',\n",
       " 'blow',\n",
       " 'blu',\n",
       " 'blue',\n",
       " 'bluetooth',\n",
       " 'blurry',\n",
       " 'board',\n",
       " 'boat',\n",
       " 'body',\n",
       " 'bonus',\n",
       " 'book',\n",
       " 'boom',\n",
       " 'boost',\n",
       " 'boot',\n",
       " 'border',\n",
       " 'bose',\n",
       " 'both',\n",
       " 'bother',\n",
       " 'bottom',\n",
       " 'bought',\n",
       " 'bounce',\n",
       " 'bounced',\n",
       " 'bow',\n",
       " 'box',\n",
       " 'boy',\n",
       " 'boyfriend',\n",
       " 'bracket',\n",
       " 'braided',\n",
       " 'brainer',\n",
       " 'brand',\n",
       " 'branded',\n",
       " 'brands',\n",
       " 'break',\n",
       " 'breakage',\n",
       " 'breaker',\n",
       " 'breaking',\n",
       " 'breaks',\n",
       " 'breeze',\n",
       " 'brick',\n",
       " 'bright',\n",
       " 'brighter',\n",
       " 'brightness',\n",
       " 'brilliant',\n",
       " 'bring',\n",
       " 'brittle',\n",
       " 'broke',\n",
       " 'broken',\n",
       " 'brother',\n",
       " 'brought',\n",
       " 'brown',\n",
       " 'bt',\n",
       " 'btw',\n",
       " 'bubble',\n",
       " 'bubbled',\n",
       " 'bubbles',\n",
       " 'buck',\n",
       " 'bucks',\n",
       " 'bud',\n",
       " 'budget',\n",
       " 'build',\n",
       " 'building',\n",
       " 'built',\n",
       " 'bulge',\n",
       " 'bulk',\n",
       " 'bulkier',\n",
       " 'bulky',\n",
       " 'bummed',\n",
       " 'bummer',\n",
       " 'bump',\n",
       " 'bumper',\n",
       " 'bumps',\n",
       " 'bumpy',\n",
       " 'bunch',\n",
       " 'bundle',\n",
       " 'business',\n",
       " 'busy',\n",
       " 'but',\n",
       " 'button',\n",
       " 'buttons',\n",
       " 'buy',\n",
       " 'buyer',\n",
       " 'buying',\n",
       " 'by',\n",
       " 'cable',\n",
       " 'cables',\n",
       " 'call',\n",
       " 'calls',\n",
       " 'cam',\n",
       " 'came',\n",
       " 'camera',\n",
       " 'cameras',\n",
       " 'camo',\n",
       " 'camp',\n",
       " 'camping',\n",
       " 'can',\n",
       " 'cancel',\n",
       " 'cannot',\n",
       " 'canon',\n",
       " 'cant',\n",
       " 'cap',\n",
       " 'capability',\n",
       " 'capable',\n",
       " 'capacity',\n",
       " 'car',\n",
       " 'carbon',\n",
       " 'card',\n",
       " 'cardboard',\n",
       " 'cards',\n",
       " 'care',\n",
       " 'careful',\n",
       " 'carefully',\n",
       " 'carrier',\n",
       " 'carry',\n",
       " 'cars',\n",
       " 'case',\n",
       " 'caseology',\n",
       " 'cases',\n",
       " 'cash',\n",
       " 'cat',\n",
       " 'catch',\n",
       " 'cause',\n",
       " 'caution',\n",
       " 'cd',\n",
       " 'cell',\n",
       " 'cellphone',\n",
       " 'cellular',\n",
       " 'cement',\n",
       " 'center',\n",
       " 'centered',\n",
       " 'certain',\n",
       " 'certainly',\n",
       " 'certified',\n",
       " 'challenging',\n",
       " 'champ',\n",
       " 'chance',\n",
       " 'change',\n",
       " 'changed',\n",
       " 'changes',\n",
       " 'changing',\n",
       " 'charge',\n",
       " 'charged',\n",
       " 'charger',\n",
       " 'chargers',\n",
       " 'charges',\n",
       " 'charging',\n",
       " 'charm',\n",
       " 'cheap',\n",
       " 'cheaper',\n",
       " 'cheaply',\n",
       " 'check',\n",
       " 'checked',\n",
       " 'checking',\n",
       " 'cheep',\n",
       " 'children',\n",
       " 'china',\n",
       " 'chinese',\n",
       " 'chip',\n",
       " 'chipped',\n",
       " 'chipping',\n",
       " 'chips',\n",
       " 'choice',\n",
       " 'choices',\n",
       " 'choose',\n",
       " 'chose',\n",
       " 'christmas',\n",
       " 'cigarette',\n",
       " 'circle',\n",
       " 'claim',\n",
       " 'claimed',\n",
       " 'claims',\n",
       " 'clamp',\n",
       " 'clarity',\n",
       " 'clasp',\n",
       " 'class',\n",
       " 'classic',\n",
       " 'classy',\n",
       " 'clean',\n",
       " 'cleaned',\n",
       " 'cleaner',\n",
       " 'cleaning',\n",
       " 'clear',\n",
       " 'clearly',\n",
       " 'click',\n",
       " 'clicks',\n",
       " 'clip',\n",
       " 'clipped',\n",
       " 'clips',\n",
       " 'clock',\n",
       " 'close',\n",
       " 'closed',\n",
       " 'closer',\n",
       " 'closing',\n",
       " 'closure',\n",
       " 'cloth',\n",
       " 'cloudy',\n",
       " 'clumsy',\n",
       " 'clunky',\n",
       " 'clutter',\n",
       " 'cm_cr_arp_d_rvw_txt',\n",
       " 'co',\n",
       " 'coating',\n",
       " 'code',\n",
       " 'cold',\n",
       " 'collect',\n",
       " 'color',\n",
       " 'colored',\n",
       " 'colorful',\n",
       " 'coloring',\n",
       " 'colors',\n",
       " 'colour',\n",
       " 'com',\n",
       " 'combination',\n",
       " 'combined',\n",
       " 'combo',\n",
       " 'come',\n",
       " 'comes',\n",
       " 'comfort',\n",
       " 'comfortable',\n",
       " 'comfortably',\n",
       " 'coming',\n",
       " 'comment',\n",
       " 'common',\n",
       " 'communication',\n",
       " 'commuter',\n",
       " 'compact',\n",
       " 'company',\n",
       " 'comparable',\n",
       " 'compare',\n",
       " 'compared',\n",
       " 'comparison',\n",
       " 'compartment',\n",
       " 'compatible',\n",
       " 'competitor',\n",
       " 'complain',\n",
       " 'complained',\n",
       " 'complaining',\n",
       " 'complains',\n",
       " 'complaint',\n",
       " 'complaints',\n",
       " 'complement',\n",
       " 'complete',\n",
       " 'completely',\n",
       " 'complicated',\n",
       " 'compliment',\n",
       " 'compliments',\n",
       " 'compromise',\n",
       " 'computer',\n",
       " 'con',\n",
       " 'concept',\n",
       " 'concern',\n",
       " 'concerned',\n",
       " 'concrete',\n",
       " 'condition',\n",
       " 'confidence',\n",
       " 'confident',\n",
       " 'connect',\n",
       " 'connected',\n",
       " 'connecting',\n",
       " 'connection',\n",
       " 'connectivity',\n",
       " 'connector',\n",
       " 'connects',\n",
       " 'cons',\n",
       " 'consider',\n",
       " 'considering',\n",
       " 'consistent',\n",
       " 'consistently',\n",
       " 'console',\n",
       " 'constant',\n",
       " 'constantly',\n",
       " 'constructed',\n",
       " 'construction',\n",
       " 'contact',\n",
       " 'contacted',\n",
       " 'continue',\n",
       " 'continued',\n",
       " 'continues',\n",
       " 'contract',\n",
       " 'control',\n",
       " 'controller',\n",
       " 'convenience',\n",
       " 'convenient',\n",
       " 'cool',\n",
       " 'coral',\n",
       " 'cord',\n",
       " 'cords',\n",
       " 'core',\n",
       " 'corner',\n",
       " 'corners',\n",
       " 'correct',\n",
       " 'correctly',\n",
       " 'cost',\n",
       " 'could',\n",
       " 'count',\n",
       " 'counter',\n",
       " 'countless',\n",
       " 'country',\n",
       " 'couple',\n",
       " 'course',\n",
       " 'cover',\n",
       " 'coverage',\n",
       " 'covered',\n",
       " 'covering',\n",
       " 'covers',\n",
       " 'crack',\n",
       " 'cracked',\n",
       " 'cracking',\n",
       " 'cracks',\n",
       " 'cradle',\n",
       " 'crap',\n",
       " 'crappy',\n",
       " 'crash',\n",
       " 'crazy',\n",
       " 'create',\n",
       " 'credit',\n",
       " 'crisp',\n",
       " 'cruzerlite',\n",
       " 'crystal',\n",
       " 'cup',\n",
       " 'current',\n",
       " 'currently',\n",
       " 'curve',\n",
       " 'curved',\n",
       " 'cushion',\n",
       " 'custom',\n",
       " 'customer',\n",
       " 'customers',\n",
       " 'cut',\n",
       " 'cute',\n",
       " 'cutout',\n",
       " 'cutouts',\n",
       " 'cutter',\n",
       " 'cutting',\n",
       " 'cuz',\n",
       " 'cycle',\n",
       " 'dad',\n",
       " 'daily',\n",
       " 'damage',\n",
       " 'damaged',\n",
       " 'damn',\n",
       " 'dark',\n",
       " 'darker',\n",
       " 'darn',\n",
       " 'dash',\n",
       " 'dashboard',\n",
       " 'data',\n",
       " 'date',\n",
       " 'daughter',\n",
       " 'daughters',\n",
       " 'day',\n",
       " 'days',\n",
       " 'dead',\n",
       " 'deal',\n",
       " 'debit',\n",
       " 'debris',\n",
       " 'decent',\n",
       " 'decide',\n",
       " 'decided',\n",
       " 'decorative',\n",
       " 'deep',\n",
       " 'def',\n",
       " 'defect',\n",
       " 'defective',\n",
       " 'defender',\n",
       " 'definately',\n",
       " 'definitely',\n",
       " 'degree',\n",
       " 'degrees',\n",
       " 'delicate',\n",
       " 'deliver',\n",
       " 'delivered',\n",
       " 'delivers',\n",
       " 'delivery',\n",
       " 'dent',\n",
       " 'depending',\n",
       " 'describe',\n",
       " 'described',\n",
       " 'description',\n",
       " 'deserves',\n",
       " 'design',\n",
       " 'designed',\n",
       " 'desired',\n",
       " 'desk',\n",
       " 'despite',\n",
       " 'destroyed',\n",
       " 'detach',\n",
       " 'detached',\n",
       " 'detail',\n",
       " 'detailed',\n",
       " 'developed',\n",
       " 'device',\n",
       " 'devices',\n",
       " 'diamond',\n",
       " 'did',\n",
       " 'didn',\n",
       " 'didnt',\n",
       " 'die',\n",
       " 'died',\n",
       " 'difference',\n",
       " 'different',\n",
       " 'difficult',\n",
       " 'difficulty',\n",
       " 'dig',\n",
       " 'digital',\n",
       " 'digitizer',\n",
       " 'dimension',\n",
       " 'ding',\n",
       " 'dings',\n",
       " 'direct',\n",
       " 'direction',\n",
       " 'directions',\n",
       " 'directly',\n",
       " 'dirt',\n",
       " 'dirty',\n",
       " 'disappear',\n",
       " 'disappoint',\n",
       " 'disappointed',\n",
       " 'disappointing',\n",
       " 'disappointment',\n",
       " 'discharge',\n",
       " 'discolored',\n",
       " 'disconnect',\n",
       " 'discount',\n",
       " 'discounted',\n",
       " 'discovered',\n",
       " 'dislike',\n",
       " 'display',\n",
       " 'displayed',\n",
       " 'dissapointed',\n",
       " 'distance',\n",
       " 'div',\n",
       " 'diztronic',\n",
       " 'do',\n",
       " 'dock',\n",
       " 'docked',\n",
       " 'docking',\n",
       " 'does',\n",
       " 'doesn',\n",
       " 'doesnt',\n",
       " 'dog',\n",
       " 'dollar',\n",
       " 'dollars',\n",
       " 'don',\n",
       " 'done',\n",
       " 'dont',\n",
       " 'door',\n",
       " 'dose',\n",
       " 'dot',\n",
       " 'double',\n",
       " 'doubt',\n",
       " 'down',\n",
       " 'downfall',\n",
       " 'download',\n",
       " 'downside',\n",
       " 'dozen',\n",
       " 'dp',\n",
       " 'drag',\n",
       " 'drain',\n",
       " 'draw',\n",
       " 'drawback',\n",
       " 'drawer',\n",
       " 'drive',\n",
       " 'driver',\n",
       " 'driving',\n",
       " 'droid',\n",
       " 'drop',\n",
       " 'dropped',\n",
       " 'dropping',\n",
       " 'drops',\n",
       " 'dry',\n",
       " 'dual',\n",
       " 'due',\n",
       " 'dull',\n",
       " 'dumb',\n",
       " 'durability',\n",
       " 'durable',\n",
       " 'dust',\n",
       " 'duty',\n",
       " 'dying',\n",
       " 'ear',\n",
       " 'earbuds',\n",
       " 'earlier',\n",
       " 'early',\n",
       " 'earphone',\n",
       " 'earphones',\n",
       " 'earpiece',\n",
       " 'ears',\n",
       " 'ease',\n",
       " 'easier',\n",
       " 'easiest',\n",
       " 'easily',\n",
       " 'easy',\n",
       " 'eat',\n",
       " 'ebay',\n",
       " 'edge',\n",
       " 'edges',\n",
       " 'edit',\n",
       " 'edition',\n",
       " 'effect',\n",
       " 'effective',\n",
       " 'effectively',\n",
       " 'efficient',\n",
       " 'effort',\n",
       " 'eight',\n",
       " 'either',\n",
       " 'el',\n",
       " 'elago',\n",
       " 'electrical',\n",
       " 'electronic',\n",
       " 'electronics',\n",
       " 'elegant',\n",
       " 'else',\n",
       " 'elsewhere',\n",
       " 'em',\n",
       " 'email',\n",
       " 'emails',\n",
       " 'emergency',\n",
       " 'enclosed',\n",
       " 'end',\n",
       " 'ended',\n",
       " 'ends',\n",
       " 'energy',\n",
       " 'english',\n",
       " 'enjoy',\n",
       " 'enjoyed',\n",
       " 'enjoying',\n",
       " 'enough',\n",
       " 'enter',\n",
       " 'entire',\n",
       " 'entirely',\n",
       " 'envelope',\n",
       " 'environment',\n",
       " 'equal',\n",
       " 'equally',\n",
       " 'equipment',\n",
       " 'error',\n",
       " 'especially',\n",
       " 'etc',\n",
       " 'even',\n",
       " 'event',\n",
       " 'eventually',\n",
       " 'ever',\n",
       " 'every',\n",
       " 'everybody',\n",
       " 'everyday',\n",
       " 'everyone',\n",
       " 'everything',\n",
       " 'everytime',\n",
       " 'everywhere',\n",
       " 'evo',\n",
       " 'exact',\n",
       " 'exactly',\n",
       " 'example',\n",
       " 'exceeded',\n",
       " 'excelente',\n",
       " 'excellent',\n",
       " 'except',\n",
       " 'exception',\n",
       " 'exchange',\n",
       " 'excited',\n",
       " 'exercise',\n",
       " 'existing',\n",
       " 'expect',\n",
       " 'expectation',\n",
       " 'expectations',\n",
       " 'expected',\n",
       " 'expecting',\n",
       " 'expensive',\n",
       " 'experience',\n",
       " 'experienced',\n",
       " 'exposed',\n",
       " 'extend',\n",
       " 'extended',\n",
       " 'extender',\n",
       " 'extends',\n",
       " 'extension',\n",
       " 'exterior',\n",
       " 'external',\n",
       " 'extra',\n",
       " 'extreme',\n",
       " 'extremely',\n",
       " 'eye',\n",
       " 'fabric',\n",
       " 'fabulous',\n",
       " 'face',\n",
       " 'facebook',\n",
       " 'fact',\n",
       " 'factor',\n",
       " 'factory',\n",
       " 'fade',\n",
       " 'fail',\n",
       " 'failed',\n",
       " 'failure',\n",
       " 'fair',\n",
       " 'fairly',\n",
       " 'fake',\n",
       " 'fall',\n",
       " 'falls',\n",
       " 'false',\n",
       " 'family',\n",
       " 'fan',\n",
       " 'fancy',\n",
       " 'fantastic',\n",
       " 'far',\n",
       " 'fashion',\n",
       " 'fashionable',\n",
       " 'fast',\n",
       " 'faster',\n",
       " 'fat',\n",
       " 'father',\n",
       " 'fault',\n",
       " 'favor',\n",
       " 'favorite',\n",
       " 'fear',\n",
       " 'feature',\n",
       " 'features',\n",
       " 'feedback',\n",
       " 'feel',\n",
       " 'feeling',\n",
       " 'feels',\n",
       " 'feet',\n",
       " 'fell',\n",
       " 'felt',\n",
       " 'feminine',\n",
       " 'fi',\n",
       " 'fiance',\n",
       " 'fiber',\n",
       " 'field',\n",
       " 'fighting',\n",
       " 'figure',\n",
       " 'figured',\n",
       " 'file',\n",
       " 'fill',\n",
       " 'film',\n",
       " 'finally',\n",
       " 'find',\n",
       " 'finding',\n",
       " 'fine',\n",
       " 'finger',\n",
       " 'fingerprint',\n",
       " 'fingerprints',\n",
       " 'fingers',\n",
       " 'finish',\n",
       " 'finished',\n",
       " 'fire',\n",
       " 'firm',\n",
       " 'firmly',\n",
       " 'first',\n",
       " 'fish',\n",
       " 'fishing',\n",
       " 'fit',\n",
       " 'fitbit',\n",
       " 'fitment',\n",
       " 'fits',\n",
       " 'fitted',\n",
       " 'fitting',\n",
       " 'five',\n",
       " 'fix',\n",
       " 'fixed',\n",
       " 'fixing',\n",
       " 'flap',\n",
       " 'flash',\n",
       " 'flashlight',\n",
       " 'flat',\n",
       " 'flaw',\n",
       " 'flawless',\n",
       " 'flawlessly',\n",
       " 'flaws',\n",
       " 'flex',\n",
       " 'flexibility',\n",
       " ...]"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y=count_vec.get_feature_names()\n",
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pydotplus\n",
    "from sklearn.tree import export_graphviz\n",
    "dot_data= export_graphviz(Dt_clf_uni,out_file=None,feature_names=y,class_names=[\"positive\",\"negative\"])\n",
    "graph = pydotplus.graph_from_dot_data(dot_data)\n",
    "graph.write_pdf(\"Dt_clf_uni_PDF\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pydotplus\n",
    "from sklearn.tree import export_graphviz\n",
    "dot_data= export_graphviz(Dt_clf_bi,out_file=None)\n",
    "graph = pydotplus.graph_from_dot_data(dot_data)\n",
    "graph.write_pdf(\"Dt_clf_bi_PDF\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Navie bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8528"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.naive_bayes import MultinomialNB\n",
    "MultinomialNB_uni_clf = MultinomialNB()\n",
    "MultinomialNB_uni_clf.fit(x_train_features,y_train)\n",
    "MultinomialNB_uni_clf.score(x_test_features,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1551,  801],\n",
       "       [ 671, 6977]], dtype=int64)"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred_uni=MultinomialNB_uni_clf.predict(x_test_features)\n",
    "cm=confusion_matrix(y_test,y_pred_uni)\n",
    "cm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         neg       0.70      0.66      0.68      2352\n",
      "         pos       0.90      0.91      0.90      7648\n",
      "\n",
      "    accuracy                           0.85     10000\n",
      "   macro avg       0.80      0.79      0.79     10000\n",
      "weighted avg       0.85      0.85      0.85     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,y_pred_uni))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Review: The phone is not running smoothly, it lags very much. Also the camera quality is bad.\n",
      "['pos']\n"
     ]
    }
   ],
   "source": [
    "Review = input(\"Review: \")\n",
    "print(MultinomialNB_uni_clf.predict(predict_uni_trans(Review)))\n",
    "#I am really happy with the awesome product quality it works really well"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8538"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "MultinomialNB_bigram_clf = MultinomialNB()\n",
    "MultinomialNB_bigram_clf.fit(x_train_features_bigram,y_train)\n",
    "MultinomialNB_bigram_clf.score(x_test_features_bigram,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1714,  638],\n",
       "       [ 824, 6824]], dtype=int64)"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred_bi=MultinomialNB_bigram_clf.predict(x_test_features_bigram)\n",
    "cm=confusion_matrix(y_test,y_pred_bi)\n",
    "cm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         neg       0.68      0.73      0.70      2352\n",
      "         pos       0.91      0.89      0.90      7648\n",
      "\n",
      "    accuracy                           0.85     10000\n",
      "   macro avg       0.79      0.81      0.80     10000\n",
      "weighted avg       0.86      0.85      0.86     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,y_pred_bi))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Review: The phone is not running smoothly, it lags very much. Also the camera quality is bad.\n",
      "['pos']\n"
     ]
    }
   ],
   "source": [
    "Review = input(\"Review: \")\n",
    "print(MultinomialNB_bigram_clf.predict(predict_bi_trans(Review)))\n",
    "#I am really happy with the awesome product quality it works really well"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('MultinomialNB_uni_countvectorize_85.28_(uniigram)_model','wb') as f:\n",
    "    pickle.dump(MultinomialNB_uni_clf,f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('MultinomialNB_bigram_countvectorize_85.38_(uniigram)_model','wb') as f:\n",
    "    pickle.dump(MultinomialNB_bigram_clf,f)"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Amazon_review_Collab.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}