Hate-Speech Analysis.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Tesina Data Spaces.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyOdo+KSkFNw4yvc4znYSeha",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/SimoneDutto/Hate-Speech-TwitterBot/blob/master/Hate-Speech%20Analysis.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oiPo5jQrReE6",
        "colab_type": "text"
      },
      "source": [
        "## Tesina Data Spaces\n",
        "\n",
        "> Indented block\n",
        "\n",
        "\n",
        "Considerazioni: stopword, lemmatize, hashtag, multinaive, logistic regression, grammar error"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7wjGZcClKEdR",
        "colab_type": "text"
      },
      "source": [
        "\n",
        "\n",
        "```\n",
        "# This is formatted as code\n",
        "```\n",
        "\n",
        "Logistic Regression, Naive Bayes, Random Forest\n",
        "*   Introduction\n",
        "*   Dataset\n",
        "*   Preprocessing\n",
        "    *   Lemmatize\n",
        "    *   Hashtag analysis\n",
        "    *   Mentions\n",
        "    *   Grammar Error\n",
        "    *   Upsampling (x)\n",
        "*   Models\n",
        "    *   Logistic Regression\n",
        "    *   Random Forest\n",
        "    *   Naive Bayes\n",
        "*   Experiment\n",
        "    *   Results\n",
        "*   Conclusion\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JxNlCgNlUQGg",
        "colab_type": "text"
      },
      "source": [
        "### Import Dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VNLSxYQ2k0Dx",
        "colab_type": "code",
        "outputId": "7769ad83-7887-45cf-d9e3-97fcb5f6ed66",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 565
        }
      },
      "source": [
        "%pip install --user -U nltk\n",
        "%pip install --user gingerit"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting nltk\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/f6/1d/d925cfb4f324ede997f6d47bea4d9babba51b49e87a767c170b77005889d/nltk-3.4.5.zip (1.5MB)\n",
            "\u001b[K     |████████████████████████████████| 1.5MB 4.8MB/s \n",
            "\u001b[?25hRequirement already satisfied, skipping upgrade: six in /usr/local/lib/python3.6/dist-packages (from nltk) (1.12.0)\n",
            "Building wheels for collected packages: nltk\n",
            "  Building wheel for nltk (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for nltk: filename=nltk-3.4.5-cp36-none-any.whl size=1449904 sha256=b01d9094b36c5863595a115cc55ab59fa97b8caaab3677cacc9155928c1a66c7\n",
            "  Stored in directory: /root/.cache/pip/wheels/96/86/f6/68ab24c23f207c0077381a5e3904b2815136b879538a24b483\n",
            "Successfully built nltk\n",
            "Installing collected packages: nltk\n",
            "Successfully installed nltk-3.4.5\n",
            "Collecting gingerit\n",
            "  Downloading https://files.pythonhosted.org/packages/78/af/e537e83cbd83b60ca94530a42a0ad9b389bad50c98b0518118e653475eeb/gingerit-0.8.0-py3-none-any.whl\n",
            "Collecting requests<3.0,>=2.22\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/1a/70/1935c770cb3be6e3a8b78ced23d7e0f3b187f5cbfab4749523ed65d7c9b1/requests-2.23.0-py2.py3-none-any.whl (58kB)\n",
            "\u001b[K     |████████████████████████████████| 61kB 1.7MB/s \n",
            "\u001b[?25hRequirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests<3.0,>=2.22->gingerit) (2019.11.28)\n",
            "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests<3.0,>=2.22->gingerit) (1.24.3)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests<3.0,>=2.22->gingerit) (3.0.4)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests<3.0,>=2.22->gingerit) (2.8)\n",
            "\u001b[31mERROR: google-colab 1.0.0 has requirement requests~=2.21.0, but you'll have requests 2.23.0 which is incompatible.\u001b[0m\n",
            "\u001b[31mERROR: datascience 0.10.6 has requirement folium==0.2.1, but you'll have folium 0.8.3 which is incompatible.\u001b[0m\n",
            "Installing collected packages: requests, gingerit\n",
            "Successfully installed gingerit-0.8.0 requests-2.23.0\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.colab-display-data+json": {
              "pip_warning": {
                "packages": [
                  "requests"
                ]
              }
            }
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "B4KC5oOgRZH_",
        "colab_type": "code",
        "outputId": "c1ad9a57-f1b5-4e40-9f4d-db1b90ea467c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 563
        }
      },
      "source": [
        "from google.colab import drive\n",
        "#link to download the dataset\n",
        "#https://www.kaggle.com/arkhoshghalb/twitter-sentiment-analysis-hatred-speech\n",
        "\n",
        "drive.mount('/content/gdrive')\n",
        "!7z x /content/gdrive/My\\ Drive/Data\\ Spaces/twitter-sentiment-analysis-hatred-speech.zip \n"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount(\"/content/gdrive\", force_remount=True).\n",
            "\n",
            "7-Zip [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21\n",
            "p7zip Version 16.02 (locale=en_US.UTF-8,Utf16=on,HugeFiles=on,64 bits,2 CPUs Intel(R) Xeon(R) CPU @ 2.20GHz (406F0),ASM,AES-NI)\n",
            "\n",
            "Scanning the drive for archives:\n",
            "  0M Scan /content/gdrive/My Drive/Data Spaces/\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b                                               \b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b1 file, 1982527 bytes (1937 KiB)\n",
            "\n",
            "Extracting archive: /content/gdrive/My Drive/Data Spaces/twitter-sentiment-analysis-hatred-speech.zip\n",
            "--\n",
            "Path = /content/gdrive/My Drive/Data Spaces/twitter-sentiment-analysis-hatred-speech.zip\n",
            "Type = zip\n",
            "Physical Size = 1982527\n",
            "\n",
            "  0%\b\b\b\b    \b\b\b\b\n",
            "Would you like to replace the existing file:\n",
            "  Path:     ./test.csv\n",
            "  Size:     1635543 bytes (1598 KiB)\n",
            "  Modified: 2019-10-17 08:06:28\n",
            "with the file from archive:\n",
            "  Path:     test.csv\n",
            "  Size:     1635543 bytes (1598 KiB)\n",
            "  Modified: 2019-10-17 08:06:28\n",
            "? (Y)es / (N)o / (A)lways / (S)kip all / A(u)to rename all / (Q)uit? s\n",
            "\n",
            " 34%\b\b\b\b    \b\b\b\bEverything is Ok\n",
            "\n",
            "Files: 0\n",
            "Size:       0\n",
            "Compressed: 1982527\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QmQqA96g6b6_",
        "colab_type": "text"
      },
      "source": [
        "## Preprocessing\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Azh8uj_d6baY",
        "colab_type": "code",
        "outputId": "d65625f7-417f-42c2-fef2-cd9e75bd3462",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 90
        }
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "df1 = pd.read_csv('train.csv')\n",
        "df_test = pd.read_csv('test.csv')\n",
        "\n",
        "print(\"Hatred labeled: {}\\nNon-hatred labeled: {}\".format(\n",
        "    (df1.label == 1).sum(),\n",
        "    (df1.label == 0).sum()\n",
        "))\n",
        "#Missing Values If Any\n",
        "df1.isna().sum()\n",
        "\n",
        "print(df1.iloc[:1, 0:3].head())\n",
        "\n"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Hatred labeled: 2242\n",
            "Non-hatred labeled: 29720\n",
            "   id  label                                              tweet\n",
            "0   1      0   @user when a father is dysfunctional and is s...\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FQUpExoxZLyQ",
        "colab_type": "text"
      },
      "source": [
        "##Define normalization functions"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Dko6CQmbgWEa",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 90
        },
        "outputId": "2c074768-18d1-4b3f-fcbb-1ec8aef08215"
      },
      "source": [
        "#importing different libraries for analysis, processing and classification\n",
        "import nltk\n",
        "from sklearn import re #regular expression for text processing\n",
        "from nltk.corpus import stopwords\n",
        "from nltk.stem import WordNetLemmatizer #word stemmer class\n",
        "lemma = WordNetLemmatizer()\n",
        "# vectorizer \n",
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from sklearn.model_selection import train_test_split, GridSearchCV\n",
        "from sklearn.linear_model import LogisticRegression #classification model\n",
        "from sklearn.metrics import confusion_matrix, classification_report, f1_score # performance evaluation criteria\n",
        "#grammar-check\n",
        "#from gingerit.gingerit import GingerIt\n",
        "\n",
        "text = 'The smelt of fliwers bring back memories.'\n",
        "\n",
        "nltk.download('wordnet')\n",
        "nltk.download('stopwords')\n",
        "\n",
        "stopwords = set(nltk.corpus.stopwords.words('english'))\n",
        "stopwords.add('amp') #amp doesnt mean anything, it is just the formatting label\n",
        "\n",
        "def normalizer(tweet):\n",
        "    tweets = \" \".join(filter(lambda x: x[0]!= '@' , tweet.split()))\n",
        "    tweets = re.sub('[^a-zA-Z]', ' ', tweets)\n",
        "    tweets = tweets.lower()\n",
        "    tweets = tweets.split()\n",
        "    tweets = [word for word in tweets if not word in stopwords]\n",
        "    tweets = [lemma.lemmatize(word) for word in tweets]\n",
        "    tweets = \" \".join(tweets)\n",
        "    return tweets\n",
        "\n",
        "def extract_hashtag(tweet):\n",
        "    tweets = \" \".join(filter(lambda x: x[0]== '#', tweet.split()))\n",
        "    tweets = re.sub('[^a-zA-Z]',' ',  tweets)\n",
        "    tweets = tweets.lower()\n",
        "    tweets = [lemma.lemmatize(word) for word in tweets]\n",
        "    tweets = \"\".join(tweets)\n",
        "    return tweets\n",
        "def mentions(tweet):\n",
        "    tweets = \" \".join(filter(lambda x: x[0]== '@' , tweet.split()))\n",
        "    return len(tweets)\n",
        "\n",
        "# def grammar_check(tweet):\n",
        "#     tweets = \" \".join(filter(lambda x: x[0]!= '@' or x[0]!='#' , tweet.split()))\n",
        "#     tweets = re.sub('[^a-zA-Z]', ' ', tweets)\n",
        "#     tweets = tweets.lower()\n",
        "#     tweets = tweets.split()\n",
        "#     tweets = [word for word in tweets if not word in stopwords]\n",
        "#     tweets = \" \".join(tweets)\n",
        "#     #print(tweets)\n",
        "#     parser = GingerIt()\n",
        "#     ginger_grammar_results = parser.parse(tweets)\n",
        "#     #pprint.pprint(ginger_grammar_results)\n",
        "#     ginger_corrections = ginger_grammar_results['corrections']\n",
        "#     return int(len(ginger_corrections))\n"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package wordnet to /root/nltk_data...\n",
            "[nltk_data]   Package wordnet is already up-to-date!\n",
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Package stopwords is already up-to-date!\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sg8sF7-9ZSBa",
        "colab_type": "text"
      },
      "source": [
        "##Apply normalization functions"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AdsPGQsCgk1I",
        "colab_type": "code",
        "outputId": "cc95331d-d188-43bc-8b95-746d34c91e49",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "source": [
        "df1['normalized_text'] = df1.tweet.apply(normalizer)\n",
        "df1['hashtag'] = df1.tweet.apply(extract_hashtag)\n",
        "df1['mentions'] = df1.tweet.apply(mentions)\n",
        "#df1['grammar_check'] = df1.iloc[:1000].tweet.apply(grammar_check)\n",
        "df1['grammar_check'] = pd.read_csv('/content/gdrive/My Drive/Data Spaces/output.csv')['grammar_check']\n",
        "# print(df1['normalized_text'].head)\n",
        "# print(df1['hashtag'].head())\n",
        "#print(df1['mentions'].head())\n",
        "#print(df1['grammar_check'].head())\n",
        "#df1.to_csv('/content/gdrive/My Drive/Data Spaces/', columns = [\"grammar_check\"])\n",
        "# print(len(df1['normalized_text']))\n",
        "all_words = set(list(' '.join(df1.tweet.values).split()))\n",
        "normalized = set(list(' '.join(df1.normalized_text.values).split()))\n",
        "print(\"Unique words all_words: \", len(all_words))\n",
        "print(\"Unique words normalized_words: \", len(normalized))"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Unique words all_words:  67223\n",
            "Unique words normalized_words:  35127\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Y-i6Ot0FZbec",
        "colab_type": "text"
      },
      "source": [
        "##Print and plot useful information about data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R0tTuRW_oILI",
        "colab_type": "code",
        "outputId": "430f3a5f-3ceb-4fb5-d633-68e9b6a175d6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "from nltk import FreqDist \n",
        "import matplotlib.pyplot as plt\n",
        "plt.style.use('ggplot')\n",
        "\n",
        "all_hashtags = FreqDist(list(' '.join(df1.hashtag.values).split())).most_common(10)\n",
        "hatred_hashtags = FreqDist(list(' '.join(df1[df1.label==1].hashtag.values).split())).most_common(10)\n",
        "plt.figure(figsize=(14, 6))\n",
        "ax = plt.subplot(121)\n",
        "pd.DataFrame(all_hashtags, columns=['hashtag', 'Count']).set_index('hashtag').plot.barh(ax=ax, fontsize=12)\n",
        "plt.xlabel('# occurrences')\n",
        "plt.title('Hashtags in all tweets', size=13)\n",
        "ax = plt.subplot(122)\n",
        "pd.DataFrame(hatred_hashtags, columns=['hashtag', 'Count']).set_index('hashtag').plot.barh(ax=ax, fontsize=12)\n",
        "plt.xlabel('# occurrences')\n",
        "plt.ylabel('')\n",
        "plt.title('Hashtags in hatred tweets', size=13)\n",
        "plt.savefig(\"hashtag.png\")\n",
        "plt.show()\n",
        "print(\"\\n\")\n",
        "all_words = FreqDist(list(' '.join(df1.normalized_text.values).split())).most_common(20)\n",
        "hatred_words = FreqDist(list(' '.join(df1[df1.label==1].normalized_text.values).split())).most_common(20)\n",
        "plt.figure(figsize=(14, 6))\n",
        "ax = plt.subplot(121)\n",
        "pd.DataFrame(all_words, columns=['word', 'Count']).set_index('word').plot.barh(ax=ax, fontsize=12)\n",
        "plt.xlabel('# occurrences')\n",
        "plt.title('Words in all tweets', size=13)\n",
        "ax = plt.subplot(122)\n",
        "pd.DataFrame(hatred_words, columns=['word', 'Count']).set_index('word').plot.barh(ax=ax, fontsize=12)\n",
        "plt.xlabel('# occurrences')\n",
        "plt.ylabel('')\n",
        "plt.title('Words in hatred tweets', size=13)\n",
        "plt.savefig(\"allwords.png\")\n",
        "plt.show()\n",
        "all_words = FreqDist(list(' '.join(df1.tweet.values).split())).most_common(20)\n",
        "hatred_words = FreqDist(list(' '.join(df1[df1.label==1].tweet.values).split())).most_common(20)\n",
        "plt.figure(figsize=(14, 6))\n",
        "ax = plt.subplot(121)\n",
        "pd.DataFrame(all_words, columns=['word', 'Count']).set_index('word').plot.barh(ax=ax, fontsize=12)\n",
        "plt.xlabel('# occurrences')\n",
        "plt.title('Words in all tweets', size=13)\n",
        "ax = plt.subplot(122)\n",
        "pd.DataFrame(hatred_words, columns=['word', 'Count']).set_index('word').plot.barh(ax=ax, fontsize=12)\n",
        "plt.xlabel('# occurrences')\n",
        "plt.ylabel('')\n",
        "plt.title('Words in hatred tweets', size=13)\n",
        "plt.savefig(\"allwords_nonorm.png\")\n",
        "plt.show()\n",
        "print(\"Number of mentions: {}\\nNumber of tweets having a mention: {}\\nCorrelation with label: {}\".format(\n",
        "    df1.mentions.sum(),\n",
        "    len(df1[df1.mentions > 0]),\n",
        "    np.corrcoef(df1.mentions, df1.label)[0][1]\n",
        "))\n",
        "#print(df1['grammar_check'].iloc[:100])\n",
        "print(\"Number of grammar errors: {}\\nNumber of tweets having a grammar error: {}\\nCorrelation with label: {}\".format(\n",
        "    df1.grammar_check.iloc[:1000].sum(),\n",
        "    len(df1[df1.grammar_check > 0]),\n",
        "    np.corrcoef(df1.grammar_check.iloc[:1000], df1.label.iloc[:1000])[0][1]\n",
        "))\n",
        "# non-relavant\n",
        "df1.drop('mentions', axis=1, inplace=True)\n",
        "df1.drop('grammar_check', axis=1, inplace=True)"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA28AAAGJCAYAAAAdTQtKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTOD/AEKfaCDySRF7AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1008x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2UAAAGJCAYAAAAKb5uDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nREREEmJTRkREREREJKH/ATAVpWWIAR/9AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1008x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/matplotlib/backends/backend_agg.py:214: RuntimeWarning: Glyph 128 missing from current font.\n",
            "  font.set_text(s, 0.0, flags=flags)\n",
            "/usr/local/lib/python3.6/dist-packages/matplotlib/backends/backend_agg.py:183: RuntimeWarning: Glyph 128 missing from current font.\n",
            "  font.set_text(s, 0, flags=flags)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1kAAAGJCAYAAAB4oPk1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1008x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Number of mentions: 93459\n",
            "Number of tweets having a mention: 11555\n",
            "Correlation with label: 0.07292294681379592\n",
            "Number of grammar errors: 2448.0\n",
            "Number of tweets having a grammar error: 996\n",
            "Correlation with label: -0.007550282679596186\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XWFalxfEZeKM",
        "colab_type": "text"
      },
      "source": [
        "##Define dataset train and test"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yRPS316kDNSI",
        "colab_type": "code",
        "outputId": "2b11c757-cd18-4280-93dd-07751ef31c4e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 163
        }
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.feature_extraction.text import CountVectorizer,TfidfVectorizer\n",
        "\n",
        "\n",
        "trainp=df1[df1['label']==1]\n",
        "trainn=df1[df1['label']==0]\n",
        "#print(trainp.info())\n",
        "#trainn.info()\n",
        "\n",
        "train_imbalanced = df1\n",
        "from sklearn.utils import resample\n",
        "df_majority = df1[df1.label==0]\n",
        "df_minority = df1[df1.label==1]\n",
        " \n",
        "# Upsample minority class\n",
        "df_minority_upsampled = resample(df_minority, \n",
        "                                 replace=True,     # sample with replacement\n",
        "                                 n_samples=len(df_majority),    # to match majority class\n",
        "                                 random_state=123) # reproducible results\n",
        "# Combine majority class with upsampled minority class\n",
        "df_upsampled = pd.concat([df_majority, df_minority_upsampled])\n",
        " \n",
        "# Display new class counts\n",
        "print(\"Before\")\n",
        "print(df1.label.value_counts())\n",
        "print(\"After\")\n",
        "print(df_upsampled.label.value_counts())\n",
        "X_train, X_test, y_train, y_test = train_test_split(df_upsampled['normalized_text'], df_upsampled['label'], test_size=0.5, stratify=df_upsampled['label'])\n",
        "\n",
        "vect = CountVectorizer()\n",
        "tf_train=vect.fit_transform(X_train)  #train the vectorizer, build the vocablury\n",
        "tf_test=vect.transform(X_test)\n",
        "#print(tf_test)\n",
        "\n",
        "# df_test['normalized_text'] = df_test.tweet.apply(normalizer)\n",
        "# tf_test_nolabel=vect.transform(df_test.normalized_text)\n",
        "tfidf = TfidfVectorizer(ngram_range=(1,3), min_df=10, stop_words=stopwords)\n",
        "\n",
        "X1_train, X1_test, y1_train, y1_test = train_test_split(df_upsampled['normalized_text'], df_upsampled['label'], test_size=0.5, stratify=df_upsampled['label'])\n",
        "tf1_train=tfidf.fit_transform(X1_train)  #train the vectorizer, build the vocablury\n",
        "tf1_test=tfidf.transform(X1_test)"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Before\n",
            "0    29720\n",
            "1     2242\n",
            "Name: label, dtype: int64\n",
            "After\n",
            "1    29720\n",
            "0    29720\n",
            "Name: label, dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Io6Ex0XYZg_y",
        "colab_type": "text"
      },
      "source": [
        "## Initial analysis on different models"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1xgl1UhDbByo",
        "colab_type": "code",
        "outputId": "06dd3216-4d02-4733-f332-dfa8c7bc9f27",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 763
        }
      },
      "source": [
        "from sklearn.naive_bayes import MultinomialNB\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn import metrics\n",
        "from mlxtend.plotting import plot_confusion_matrix\n",
        "import time\n",
        "\n",
        "\n",
        "model = MultinomialNB()\n",
        "start = time.time()\n",
        "model.fit(X=tf_train,y=y_train)\n",
        "stop = time.time()\n",
        "print(f\"Training time: {stop - start}s\")\n",
        "expected = y_test\n",
        "start = time.time()\n",
        "predicted=model.predict(tf_test)\n",
        "stop = time.time()\n",
        "print(f\"Test time: {stop - start}s\")\n",
        "print(metrics.classification_report(expected, predicted))\n",
        "#print(metrics.confusion_matrix(expected, predicted))\n",
        "\n",
        "#plot_confusion_matrix(metrics.confusion_matrix(expected, predicted))\n",
        "model_1 = LogisticRegression(C=10)\n",
        "start = time.time()\n",
        "model_1.fit(X=tf1_train,y=y1_train)\n",
        "stop = time.time()\n",
        "print(f\"Training time: {stop - start}s\")\n",
        "expected1 = y1_test\n",
        "start = time.time()\n",
        "predicted1=model_1.predict(tf1_test)\n",
        "stop = time.time()\n",
        "print(f\"Test time: {stop - start}s\")\n",
        "print(metrics.classification_report(expected1, predicted1))\n",
        "\n",
        "model_2 = RandomForestClassifier()\n",
        "start = time.time()\n",
        "model_2.fit(X=tf1_train,y=y1_train)\n",
        "stop = time.time()\n",
        "print(f\"Training time: {stop - start}s\")\n",
        "expected2 = y1_test\n",
        "start = time.time()\n",
        "predicted2=model_2.predict(tf1_test)\n",
        "stop = time.time()\n",
        "print(f\"Test time: {stop - start}s\")\n",
        "print(metrics.classification_report(expected2, predicted2))"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Training time: 0.010847330093383789s\n",
            "Test time: 0.003029346466064453s\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.98      0.92      0.95     14860\n",
            "           1       0.92      0.98      0.95     14860\n",
            "\n",
            "    accuracy                           0.95     29720\n",
            "   macro avg       0.95      0.95      0.95     29720\n",
            "weighted avg       0.95      0.95      0.95     29720\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
            "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
            "\n",
            "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
            "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
            "Please also refer to the documentation for alternative solver options:\n",
            "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
            "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Training time: 0.5560946464538574s\n",
            "Test time: 0.0011341571807861328s\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      0.95      0.97     14860\n",
            "           1       0.95      0.99      0.97     14860\n",
            "\n",
            "    accuracy                           0.97     29720\n",
            "   macro avg       0.97      0.97      0.97     29720\n",
            "weighted avg       0.97      0.97      0.97     29720\n",
            "\n",
            "Training time: 16.4473774433136s\n",
            "Test time: 2.2480313777923584s\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       1.00      0.98      0.99     14860\n",
            "           1       0.98      1.00      0.99     14860\n",
            "\n",
            "    accuracy                           0.99     29720\n",
            "   macro avg       0.99      0.99      0.99     29720\n",
            "weighted avg       0.99      0.99      0.99     29720\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3xt9Pnp7EfbG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 591
        },
        "outputId": "41f1c7ac-ab1a-4674-a622-1669f3d3dc4c"
      },
      "source": [
        "model_1 = LogisticRegression(C=10)\n",
        "model_1.fit(X=tf1_train,y=y1_train)\n",
        "expected1 = y1_test\n",
        "predicted1=model_1.predict(tf1_test)\n",
        "predicted1_proba = model_1.predict_proba(tf1_test)\n",
        "print(metrics.classification_report(expected1, predicted1))\n",
        "threshold = np.arange(0.1,0.9,0.1)\n",
        "score = [f1_score(y1_test, ((predicted1_proba[:,1] >= x).astype(int))) for x in threshold]\n",
        "plt.plot(threshold, score)\n",
        "plt.xlabel('Threshold Probability')\n",
        "plt.ylabel('F1 score')\n",
        "plt.show()"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
            "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
            "\n",
            "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
            "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
            "Please also refer to the documentation for alternative solver options:\n",
            "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
            "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      0.95      0.97     14860\n",
            "           1       0.95      0.99      0.97     14860\n",
            "\n",
            "    accuracy                           0.97     29720\n",
            "   macro avg       0.97      0.97      0.97     29720\n",
            "weighted avg       0.97      0.97      0.97     29720\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEJCAYAAABlmAtYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MShGBbQgZndP",
        "colab_type": "text"
      },
      "source": [
        "## Gridsearch to fine tune models"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7n0YSI7zqkTJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "from matplotlib.colors import ListedColormap, Normalize\n",
        "from matplotlib.ticker import FormatStrFormatter, StrMethodFormatter\n",
        "import numpy as np\n",
        "\n",
        "def plot_grid_search(cv_results, grid_param_1, grid_param_2, name_param_1, name_param_2, log=False):\n",
        "    # Get Test Scores Mean and std for each grid search\n",
        "    scores_mean = cv_results['mean_test_score']\n",
        "    scores_mean = np.array(scores_mean).reshape(len(grid_param_2),len(grid_param_1))\n",
        "\n",
        "    scores_sd = cv_results['std_test_score']\n",
        "    scores_sd = np.array(scores_sd).reshape(len(grid_param_2),len(grid_param_1))\n",
        "\n",
        "    # Plot Grid search scores\n",
        "    _, ax = plt.subplots(1,1)\n",
        "\n",
        "    # Param1 is the X-axis, Param 2 is represented as a different curve (color line)\n",
        "    for idx, val in enumerate(grid_param_2):\n",
        "        ax.plot(grid_param_1, scores_mean[idx,:], '-o', label= name_param_2 + ': ' + str(val))\n",
        "\n",
        "    ax.set_xlabel(name_param_1, fontsize=16)\n",
        "    ax.set_ylabel('CV Average Score', fontsize=16)\n",
        "    if log:\n",
        "      plt.xscale('log')\n",
        "    ax.legend(loc='upper center', bbox_to_anchor=(1.45, 0.8), shadow=True, ncol=1)\n",
        "    ax.grid('on')\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k5GQ6GXYLHEP",
        "colab_type": "code",
        "outputId": "a39d8cc0-3ac4-4747-8e5a-e3e509a6cba2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 668
        }
      },
      "source": [
        "from sklearn.metrics import recall_score, accuracy_score\n",
        "accuracies=[]\n",
        "recalls=[]\n",
        "Cs = [0.1, 1, 10, 100, 1000]\n",
        "param_grid = {'C': Cs}\n",
        "\n",
        "clf = GridSearchCV( LogisticRegression(), param_grid = param_grid, cv = 5, verbose=True, n_jobs=-1)\n",
        "    \n",
        "clf.fit(tf1_train, y1_train)\n",
        "\n",
        "plt.plot(Cs, clf.cv_results_['mean_test_score'])\n",
        "plt.xlabel('Cs')\n",
        "plt.xscale('log')\n",
        "plt.ylabel('Average CV Accuracy')\n",
        "plt.show()\n",
        "print(\"Best model on training set has %r with accuracy %.4f\" %(clf.best_params_, clf.best_score_))\n",
        "y_true, y1_pred = y1_test, clf.predict(tf1_test)\n",
        "print(classification_report(y_true, y1_pred))\n",
        "accuracies.append(accuracy_score(y_true, y1_pred))\n",
        "recalls.append(recall_score(y_true, y1_pred))"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Fitting 5 folds for each of 5 candidates, totalling 25 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done  25 out of  25 | elapsed:    8.0s finished\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
            "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
            "\n",
            "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
            "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
            "Please also refer to the documentation for alternative solver options:\n",
            "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
            "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAENCAYAAAD+CUlOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Best model on training set has {'C': 100} with accuracy 0.9709\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      0.95      0.97     14860\n",
            "           1       0.95      0.99      0.97     14860\n",
            "\n",
            "    accuracy                           0.97     29720\n",
            "   macro avg       0.97      0.97      0.97     29720\n",
            "weighted avg       0.97      0.97      0.97     29720\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "73bBz8E66SyY",
        "colab_type": "code",
        "outputId": "6f8fb0e1-bb31-4b47-e86d-b632a3b53e52",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 541
        }
      },
      "source": [
        "n_estimators = list(range(10,101,20))\n",
        "max_features = list(range(6,32,5))\n",
        "max_depth = list(range(6,32,5))\n",
        "param_grid = {'n_estimators': n_estimators, 'max_features' : max_features}\n",
        "#param_grid = {'n_estimators': n_estimators, 'max_depth' : max_depth}\n",
        "clf_for = GridSearchCV(RandomForestClassifier(), param_grid = param_grid, cv = 5, verbose=True, n_jobs=-1)\n",
        "    \n",
        "clf_for.fit(tf1_train, y1_train)\n",
        "\n",
        "plot_grid_search(clf_for.cv_results_, n_estimators, max_features, \"n_estimators\", \"max_features\")\n",
        "#plot_grid_search(clf_for.cv_results_, n_estimators, max_depth, \"n_estimators\", \"max_depth\")\n",
        "\n",
        "print(\"Best model on training set has %r with accuracy %.4f\" %(clf.best_params_, clf.best_score_))\n",
        "y_true, y_pred = y1_test, clf_for.predict(tf1_test)\n",
        "print(classification_report(y_true, y_pred))\n",
        "accuracies.append(accuracy_score(y_true, y_pred))\n",
        "recalls.append(recall_score(y_true, y_pred))"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Fitting 5 folds for each of 30 candidates, totalling 150 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done  46 tasks      | elapsed:  4.5min\n",
            "[Parallel(n_jobs=-1)]: Done 150 out of 150 | elapsed: 14.3min finished\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Best model on training set has {'C': 100} with accuracy 0.9709\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       1.00      0.99      1.00     14860\n",
            "           1       0.99      1.00      1.00     14860\n",
            "\n",
            "    accuracy                           1.00     29720\n",
            "   macro avg       1.00      1.00      1.00     29720\n",
            "weighted avg       1.00      1.00      1.00     29720\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmgAAAENCAYAAAC7Ppk9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nl7fcMgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "opG9eijZKMir",
        "colab_type": "code",
        "outputId": "886390bf-cfe9-4fa8-d95f-089b9f7a1453",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 555
        }
      },
      "source": [
        "alphas = [0, 0.001, 0.1,  0.3, 0.5, 0.7, 1]\n",
        "\n",
        "param_grid = {'alpha': alphas}\n",
        "\n",
        "clf_nai = GridSearchCV(MultinomialNB(), param_grid = param_grid, cv = 5, verbose=True, n_jobs=-1)\n",
        "    \n",
        "clf_nai.fit(tf_train, y_train)\n",
        "\n",
        "plt.plot(alphas, clf_nai.cv_results_['mean_test_score'])\n",
        "plt.xlabel('Alpha')\n",
        "plt.ylabel('Average CV Accuracy')\n",
        "plt.show()\n",
        "\n",
        "\n",
        "print(\"Best model on training set has %r with accuracy %.4f\" %(clf_nai.best_params_, clf_nai.best_score_))\n",
        "y_true, y_pred = y_test, clf_nai.predict(tf_test)\n",
        "print(classification_report(y_true, y_pred))\n",
        "accuracies.append(accuracy_score(y_true, y_pred))\n",
        "recalls.append(recall_score(y_true, y_pred))"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Fitting 5 folds for each of 7 candidates, totalling 35 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done  35 out of  35 | elapsed:    0.5s finished\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/naive_bayes.py:507: UserWarning: alpha too small will result in numeric errors, setting alpha = 1.0e-10\n",
            "  'setting alpha = %.1e' % _ALPHA_MIN)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAAEJCAYAAACpATGzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEE7xfytZBNXoZsbrAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Best model on training set has {'alpha': 0} with accuracy 0.9609\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.98      0.95      0.97     14860\n",
            "           1       0.95      0.98      0.97     14860\n",
            "\n",
            "    accuracy                           0.97     29720\n",
            "   macro avg       0.97      0.97      0.97     29720\n",
            "weighted avg       0.97      0.97      0.97     29720\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Uoq0U-w8avkO",
        "colab_type": "code",
        "outputId": "156815e4-7b5e-4ef3-8648-52269c9e0753",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 337
        }
      },
      "source": [
        "# Setting the positions and width for the bars\n",
        "pos = list(range(len(accuracies))) \n",
        "width = 0.25 \n",
        "    \n",
        "# Plotting the bars\n",
        "fig, ax = plt.subplots(figsize=(10,5))\n",
        "\n",
        "# Create a bar with pre_score data,\n",
        "# in position pos,\n",
        "plt.bar(pos, \n",
        "        #using df['pre_score'] data,\n",
        "        accuracies, \n",
        "        # of width\n",
        "        width, \n",
        "        # with alpha 0.5\n",
        "        alpha=0.5, \n",
        "        # with color\n",
        "        color='#EE3224', \n",
        "        # with label the first value in first_name\n",
        "        label=\"Acc#uracy\") \n",
        "\n",
        "# Create a bar with mid_score data,\n",
        "# in position pos + some width buffer,\n",
        "plt.bar([p + width for p in pos], \n",
        "        #using df['mid_score'] data,\n",
        "        recalls,\n",
        "        # of width\n",
        "        width, \n",
        "        # with alpha 0.5\n",
        "        alpha=0.5, \n",
        "        # with color\n",
        "        color='#F78F1E', \n",
        "        # with label the second value in first_name\n",
        "        label=\"Recall\") \n",
        "\n",
        "# Set the y axis label\n",
        "ax.set_ylabel('% achieved on test set')\n",
        "\n",
        "# Set the chart's title\n",
        "ax.set_title('Models')\n",
        "\n",
        "# Set the position of the x ticks\n",
        "ax.set_xticks([p+ 0.5 * width for p in pos])\n",
        "\n",
        "# Set the labels for the x ticks\n",
        "ax.set_xticklabels([\"Logistic Regression\",\"Random Forest\", \"Naive-Bayes\"])\n",
        "\n",
        "# Setting the x-axis and y-axis limits\n",
        "plt.xlim(min(pos)-width, max(pos)+width*4)\n",
        "plt.ylim([0.9, max(accuracies + recalls)] )\n",
        "\n",
        "# Adding the legend and showing the plot\n",
        "plt.legend(['Accuracy', 'Recall'] ,loc='upper right')\n",
        "plt.grid()\n",
        "plt.show()"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmoAAAFACAYAAAAMDVCyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAHgoghoAAICHIqgBAAB4KIIaAACAh/p/9KpMUzKHJ3UAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 720x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}