NLP_test.ipynb

nlp_test.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "NLP_test.ipynb",
      "provenance": [],
      "toc_visible": true,
      "mount_file_id": "1SKwoeAgvJx-alJaS1HG3JqJ0i6vGjZYp",
      "authorship_tag": "ABX9TyOaIJylIUt5sYqjE1dnrbvA",
      "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/gist/akhildaphara/9c0e8fd50def1b6eee6f3eb977584417/nlp_test.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "viU2VDxie3z6"
      },
      "source": [
        "import pandas as pd"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1Bal3HEqfUE3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "outputId": "f328b6ad-5f75-4044-84be-028e508eb3d5"
      },
      "source": [
        "dataset = pd.read_csv(\"/content/drive/My Drive/Codes/NLP/Restaurant_Reviews.tsv\", delimiter='\\t')\n",
        "dataset"
      ],
      "execution_count": 2,
      "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>Liked</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Wow... Loved this place.</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Crust is not good.</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Not tasty and the texture was just nasty.</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Stopped by during the late May bank holiday of...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>The selection on the menu was great and so wer...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>995</th>\n",
              "      <td>I think food should have flavor and texture an...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>996</th>\n",
              "      <td>Appetite instantly gone.</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>997</th>\n",
              "      <td>Overall I was not impressed and would not go b...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>998</th>\n",
              "      <td>The whole experience was underwhelming, and I ...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>999</th>\n",
              "      <td>Then, as if I hadn't wasted enough of my life ...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1000 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                Review  Liked\n",
              "0                             Wow... Loved this place.      1\n",
              "1                                   Crust is not good.      0\n",
              "2            Not tasty and the texture was just nasty.      0\n",
              "3    Stopped by during the late May bank holiday of...      1\n",
              "4    The selection on the menu was great and so wer...      1\n",
              "..                                                 ...    ...\n",
              "995  I think food should have flavor and texture an...      0\n",
              "996                           Appetite instantly gone.      0\n",
              "997  Overall I was not impressed and would not go b...      0\n",
              "998  The whole experience was underwhelming, and I ...      0\n",
              "999  Then, as if I hadn't wasted enough of my life ...      0\n",
              "\n",
              "[1000 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qb6W3rYihJSP",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9969bc71-cdfc-4748-a255-dff5b4f8260e"
      },
      "source": [
        "import re\n",
        "import nltk\n",
        "nltk.download('stopwords')\n",
        "from nltk.corpus import stopwords\n",
        "from nltk.stem.porter import PorterStemmer\n",
        "\n",
        "corpus = []\n",
        "for i in range(0,1000):\n",
        "  review = re.sub('[^a-zA-Z]', ' ', dataset['Review'][i])\n",
        "  review = review.lower()\n",
        "  review = review.split()\n",
        "  ps = PorterStemmer()\n",
        "  review = [ps.stem(word) for word in review if not word in set(stopwords.words('english')) or word=='not' ]\n",
        "  review = ' '.join(review)\n",
        "  corpus.append(review)"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Unzipping corpora/stopwords.zip.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ncp_nWMBjBMu"
      },
      "source": [
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "cv = CountVectorizer(max_features = 1500)\n",
        "X = cv.fit_transform(corpus).toarray()\n",
        "y = dataset.iloc[:, 1].values"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GTw-AMqkjFqG"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 0)"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XX9KmF_OsJ1u"
      },
      "source": [
        "Random Forest"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XuJH1p0FrWO7",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5abe0419-3ba6-447c-bc32-8a3fe3d6d3da"
      },
      "source": [
        "from sklearn.ensemble import RandomForestClassifier\n",
        "classifier = RandomForestClassifier(n_estimators=1000)\n",
        "classifier.fit(X_train, y_train)\n",
        "y_pred = classifier.predict(X_test)\n",
        "classifier.score(X_test, y_test)"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.765"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FasIwD4NsIIi"
      },
      "source": [
        "SVC"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "204f6yKokJx4"
      },
      "source": [
        "from sklearn.svm import SVC\n",
        "classifier = SVC(C = 2, kernel = 'linear', random_state = 0)\n",
        "classifier.fit(X_train, y_train)\n",
        "y_pred = classifier.predict(X_test)"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fVKgW8bBh2XR"
      },
      "source": [
        "Prediction"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sDDhC3Jyh42N"
      },
      "source": [
        "def predict(new_review):    \n",
        "    new_review = re.sub(\"[^a-zA-Z]\", \" \", new_review)    \n",
        "    new_review = new_review.lower().split()\n",
        "    new_review = [ps.stem(word) for word in new_review if word not in set(stopwords.words(\"english\")) or word=='not']    \n",
        "    new_review = \" \".join(new_review)    \n",
        "    new_review = [new_review]    \n",
        "    new_review = cv.transform(new_review).toarray()    \n",
        "    if classifier.predict(new_review)[0] == 1:\n",
        "        return \"Positive\"    \n",
        "    else:        \n",
        "        return \"Negative\""
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MEVjVt2Xh_ju",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "9818d1fe-4e0c-48a8-eb59-1326c777bc0c"
      },
      "source": [
        "predict(\"Not Good food\")"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'Negative'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "52LWNdgkctrY"
      },
      "source": [
        "End of Code\n",
        "---\n",
        "\n",
        "\n",
        "Model Evaluation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IWeoae8KkbLd",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e95a23f5-fcf9-4475-ed2b-5efe46d61925"
      },
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "cm = confusion_matrix(y_test, y_pred)\n",
        "print(cm)\n",
        "print(\"Accuracy= \"+str((cm[0][0]+cm[1][1])/200))"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[78 19]\n",
            " [23 80]]\n",
            "Accuracy= 0.79\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bNcJVPMxm58k",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "034e4e75-47f9-40c9-fd2f-8fc7d637250d"
      },
      "source": [
        "classifier.score(X_test, y_test)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.79"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FoK7ue1Kqh7G",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5d4e57ec-fa61-4a5c-827f-b6d4110737b1"
      },
      "source": [
        "classifier.score(X_train, y_train)"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9875"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MjWvXrO4pXqO",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "14b612c7-b554-40e0-e3ec-c9be0a7b5e5c"
      },
      "source": [
        "from sklearn.model_selection import cross_val_score\n",
        "accuracy = cross_val_score(estimator = classifier, X= X_train, y = y_train, cv= 10)\n",
        "print(accuracy.mean())\n",
        "accuracy.std()"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.8\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.04330127018922194"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TStqeQCZqCqq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "39e4066a-fc7e-4c59-9011-9715bf7f1d14"
      },
      "source": [
        "from sklearn.model_selection import GridSearchCV\n",
        "parameters = [{'C' : [1, 2, 2.5, 3, 4], 'kernel' : ['linear']}]\n",
        "              \n",
        "grid_search = GridSearchCV(estimator= classifier, param_grid=parameters, scoring= 'accuracy', cv = 10, n_jobs = -1)\n",
        "grid_search = grid_search.fit(X_train, y_train)\n",
        "best_accuracy = grid_search.best_score_\n",
        "print(best_accuracy)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.8\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HZ_lnDy1vy8-",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e168df7d-d14c-404f-8803-0a34d1eeb6ae"
      },
      "source": [
        "best_param = grid_search.best_params_\n",
        "print(best_param)"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{'C': 2, 'kernel': 'linear'}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0VlKsg6c1XXy",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "912454d0-21fa-4078-f1cc-a75caa7bb2bf"
      },
      "source": [
        "from sklearn.metrics import classification_report\n",
        "print(classification_report(y_test, y_pred))"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.77      0.80      0.79        97\n",
            "           1       0.81      0.78      0.79       103\n",
            "\n",
            "    accuracy                           0.79       200\n",
            "   macro avg       0.79      0.79      0.79       200\n",
            "weighted avg       0.79      0.79      0.79       200\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}