SIE533_Project_English.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "cba7f800",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     C:\\Users\\harsh\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import re\n",
    "import nltk\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "from nltk.corpus import stopwords\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "nltk.download('stopwords')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2e11baa5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load contents, correlations and topics file\n",
    "BASE_DIR = \"data/\"\n",
    "content = pd.read_csv(BASE_DIR+\"content.csv\")\n",
    "correlations = pd.read_csv(BASE_DIR+\"correlations.csv\")\n",
    "topic = pd.read_csv(BASE_DIR+\"topics.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "07bd489e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original contents file shape: (154047, 8)\n",
      "Original topics file shape: (76972, 9)\n",
      "Original correlations file shape: (61517, 2)\n"
     ]
    }
   ],
   "source": [
    "# Print the original size of the data\n",
    "print(\"Original contents file shape:\", content.shape)\n",
    "print(\"Original topics file shape:\", topic.shape)\n",
    "print(\"Original correlations file shape:\", correlations.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1664281c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Use only CONTENT whose language is english\n",
    "english_filtered_content = content[content[\"language\"]==\"en\"]\n",
    "# Use only TOPICS whose language is english & has content & drop rows if there are any NAN present in the dataset\n",
    "english_filtered_topics = topic[(topic[\"language\"]==\"en\") & (topic[\"has_content\"]==1)].dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a1ed4b1f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Filtered content shape: (65939, 8)\n",
      "Filtered topic shape: (9073, 9)\n",
      "Filtered correlations shape: (61517, 2)\n"
     ]
    }
   ],
   "source": [
    "# Print the filtered size of the data\n",
    "print(\"Filtered content shape:\", english_filtered_content.shape)\n",
    "print(\"Filtered topic shape:\", english_filtered_topics.shape)\n",
    "print(\"Filtered correlations shape:\", correlations.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6b19b3ad",
   "metadata": {},
   "source": [
    "#### Filter based on content"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "765d609c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Filtered content shape: (6847, 8)\n"
     ]
    }
   ],
   "source": [
    "# We want to focus on CONTENT which is available as a document or video or html5\n",
    "english_filtered_content = english_filtered_content[(english_filtered_content[\"kind\"]==\"document\")]\n",
    "english_filtered_content = english_filtered_content.dropna()\n",
    "print(\"Filtered content shape:\", english_filtered_content.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "5181a8a3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# First find the topics which are in correlations dataset and filtered english topic dataset\n",
    "english_filtered_correlation = correlations[correlations['topic_id'].isin(english_filtered_topics['id'])]\n",
    "# print(english_filtered_correlation.shape)\n",
    "# Each topic has multiple contents inside it - so split the correlations dataset so that each topic is associated to a single content file\n",
    "exploded_english_filtered_correlation = english_filtered_correlation.assign(content_ids=english_filtered_correlation['content_ids'].str.split()).explode('content_ids')\n",
    "# print(english_filtered_correlation.shape)\n",
    "# Find the contents which are in correlations dataset and filtered english content dataset\n",
    "exploded_english_filtered_correlation = exploded_english_filtered_correlation[exploded_english_filtered_correlation['content_ids'].isin(english_filtered_content['id'])]\n",
    "# print(english_filtered_correlation.shape)\n",
    "# Count the number of content available for each topic\n",
    "new_english_filtered_correlation = exploded_english_filtered_correlation[\"topic_id\"].value_counts().to_frame('counts').reset_index()\n",
    "# english_filtered_correlation_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "7e6b02e9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            index  counts\n",
      "0  t_58ca0afc0bae     270\n",
      "1  t_17bf2af6d82b     214\n",
      "2  t_ba5de272b4cb     205\n"
     ]
    }
   ],
   "source": [
    "# Get topics that have the highest amount of content available\n",
    "top_topics = new_english_filtered_correlation.iloc[0:3]\n",
    "print(top_topics)\n",
    "first_topic, second_topic = top_topics.iloc[0,0], top_topics.iloc[1,0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "189b13f1",
   "metadata": {},
   "outputs": [],
   "source": [
    "first_topic_content = english_filtered_correlation[english_filtered_correlation['topic_id']==first_topic][\"content_ids\"]\n",
    "second_topic_content = english_filtered_correlation[english_filtered_correlation['topic_id']==second_topic][\"content_ids\"]\n",
    "# third_topic_content = english_filtered_correlation[english_filtered_correlation['topic_id']==third_topic][\"content_ids\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "f02f7f6a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "484"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "first_list = first_topic_content.str.split().tolist()\n",
    "second_list = second_topic_content.str.split().tolist()\n",
    "# third_list = third_topic_content.str.split().tolist()\n",
    "final_list = first_list[0] + second_list[0] #+ third_list[0]\n",
    "len(final_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "66677310-0a2e-44a7-84a0-c21deeb42ec5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(484, 1)"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "zeros = np.zeros((len(first_list[0]),1), dtype=int)\n",
    "ones = np.ones((len(second_list[0]),1), dtype=int)\n",
    "# twos = 2*np.ones((len(third_list[0]),1), dtype=int)\n",
    "\n",
    "result_array = np.concatenate((zeros, ones))\n",
    "result_array.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "4a202fa4",
   "metadata": {},
   "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>content_ids</th>\n",
       "      <th>true_values</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>c_0286ded417a2</td>\n",
       "      <td>[0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>c_02a5e75e01f0</td>\n",
       "      <td>[0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>c_02adc62b7731</td>\n",
       "      <td>[0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>c_04213823e40c</td>\n",
       "      <td>[0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>c_044e55056282</td>\n",
       "      <td>[0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>479</th>\n",
       "      <td>c_faaa92210fe1</td>\n",
       "      <td>[1]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>480</th>\n",
       "      <td>c_faaea6c6ad4c</td>\n",
       "      <td>[1]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>481</th>\n",
       "      <td>c_fae0b7c7a87a</td>\n",
       "      <td>[1]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>482</th>\n",
       "      <td>c_fd75a88c2bb4</td>\n",
       "      <td>[1]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>483</th>\n",
       "      <td>c_ff73c6a38f19</td>\n",
       "      <td>[1]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>484 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        content_ids true_values\n",
       "0    c_0286ded417a2         [0]\n",
       "1    c_02a5e75e01f0         [0]\n",
       "2    c_02adc62b7731         [0]\n",
       "3    c_04213823e40c         [0]\n",
       "4    c_044e55056282         [0]\n",
       "..              ...         ...\n",
       "479  c_faaa92210fe1         [1]\n",
       "480  c_faaea6c6ad4c         [1]\n",
       "481  c_fae0b7c7a87a         [1]\n",
       "482  c_fd75a88c2bb4         [1]\n",
       "483  c_ff73c6a38f19         [1]\n",
       "\n",
       "[484 rows x 2 columns]"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "curriculum_dataframe = pd.DataFrame({\"content_ids\" : final_list, \"true_values\" : result_array.tolist()})\n",
    "curriculum_dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "596eef40",
   "metadata": {},
   "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>content_ids</th>\n",
       "      <th>true_values</th>\n",
       "      <th>title</th>\n",
       "      <th>description</th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>c_0286ded417a2</td>\n",
       "      <td>[0]</td>\n",
       "      <td>Chapter 142: RenderScript</td>\n",
       "      <td>Chapter 142 of the book on Android.</td>\n",
       "      <td>Chapter 142: RenderScript\\nRenderScript is a s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>c_02a5e75e01f0</td>\n",
       "      <td>[0]</td>\n",
       "      <td>Chapter 161: TransitionDrawable</td>\n",
       "      <td>Chapter 161 of the book on Android.</td>\n",
       "      <td>Chapter 161: TransitionDrawable\\nSection 161.1...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>c_02adc62b7731</td>\n",
       "      <td>[0]</td>\n",
       "      <td>Chapter 31: Supporting Screens With Dierent R...</td>\n",
       "      <td>Chapter 31 of the book on Android.</td>\n",
       "      <td>Chapter 31: Supporting Screens With\\nDierent ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>c_04213823e40c</td>\n",
       "      <td>[0]</td>\n",
       "      <td>Chapter 93: RoboGuice</td>\n",
       "      <td>Chapter 93 of the book on Android.</td>\n",
       "      <td>Chapter 93: RoboGuice\\nSection 93.1: Simple ex...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>c_044e55056282</td>\n",
       "      <td>[0]</td>\n",
       "      <td>Chapter 219: Gson</td>\n",
       "      <td>Chapter 219 of the book on Android.</td>\n",
       "      <td>Chapter 219: Gson\\nGson is a Java library that...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>479</th>\n",
       "      <td>c_faaa92210fe1</td>\n",
       "      <td>[1]</td>\n",
       "      <td>Chapter 148: GameCenter Leaderboards</td>\n",
       "      <td>Chapter 148 of the book on iOS Developer.</td>\n",
       "      <td>Chapter 148: GameCenter Leaderboards\\nSection ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>480</th>\n",
       "      <td>c_faaea6c6ad4c</td>\n",
       "      <td>[1]</td>\n",
       "      <td>Content list</td>\n",
       "      <td>Section \"Content list\" of the book on iOS Deve...</td>\n",
       "      <td>iOS\\nDeveloper\\niOS Developer\\nNotes for Profe...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>481</th>\n",
       "      <td>c_fae0b7c7a87a</td>\n",
       "      <td>[1]</td>\n",
       "      <td>Chapter 42: UIActivityViewController</td>\n",
       "      <td>Chapter 42 of the book on iOS Developer.</td>\n",
       "      <td>Chapter 42: UIActivityViewController\\nParamete...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>482</th>\n",
       "      <td>c_fd75a88c2bb4</td>\n",
       "      <td>[1]</td>\n",
       "      <td>Chapter 25: UITextField Delegate</td>\n",
       "      <td>Chapter 25 of the book on iOS Developer.</td>\n",
       "      <td>Chapter 25: UITextField Delegate\\nSection 25.1...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>483</th>\n",
       "      <td>c_ff73c6a38f19</td>\n",
       "      <td>[1]</td>\n",
       "      <td>Chapter 166: AirPrint tutorial in iOS</td>\n",
       "      <td>Chapter 166 of the book on iOS Developer.</td>\n",
       "      <td>Chapter 166: AirPrint tutorial in iOS\\nSection...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>484 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        content_ids true_values  \\\n",
       "0    c_0286ded417a2         [0]   \n",
       "1    c_02a5e75e01f0         [0]   \n",
       "2    c_02adc62b7731         [0]   \n",
       "3    c_04213823e40c         [0]   \n",
       "4    c_044e55056282         [0]   \n",
       "..              ...         ...   \n",
       "479  c_faaa92210fe1         [1]   \n",
       "480  c_faaea6c6ad4c         [1]   \n",
       "481  c_fae0b7c7a87a         [1]   \n",
       "482  c_fd75a88c2bb4         [1]   \n",
       "483  c_ff73c6a38f19         [1]   \n",
       "\n",
       "                                                 title  \\\n",
       "0                            Chapter 142: RenderScript   \n",
       "1                      Chapter 161: TransitionDrawable   \n",
       "2    Chapter 31: Supporting Screens With Dierent R...   \n",
       "3                                Chapter 93: RoboGuice   \n",
       "4                                    Chapter 219: Gson   \n",
       "..                                                 ...   \n",
       "479               Chapter 148: GameCenter Leaderboards   \n",
       "480                                       Content list   \n",
       "481               Chapter 42: UIActivityViewController   \n",
       "482                   Chapter 25: UITextField Delegate   \n",
       "483              Chapter 166: AirPrint tutorial in iOS   \n",
       "\n",
       "                                           description  \\\n",
       "0                  Chapter 142 of the book on Android.   \n",
       "1                  Chapter 161 of the book on Android.   \n",
       "2                   Chapter 31 of the book on Android.   \n",
       "3                   Chapter 93 of the book on Android.   \n",
       "4                  Chapter 219 of the book on Android.   \n",
       "..                                                 ...   \n",
       "479          Chapter 148 of the book on iOS Developer.   \n",
       "480  Section \"Content list\" of the book on iOS Deve...   \n",
       "481           Chapter 42 of the book on iOS Developer.   \n",
       "482           Chapter 25 of the book on iOS Developer.   \n",
       "483          Chapter 166 of the book on iOS Developer.   \n",
       "\n",
       "                                                  text  \n",
       "0    Chapter 142: RenderScript\\nRenderScript is a s...  \n",
       "1    Chapter 161: TransitionDrawable\\nSection 161.1...  \n",
       "2    Chapter 31: Supporting Screens With\\nDierent ...  \n",
       "3    Chapter 93: RoboGuice\\nSection 93.1: Simple ex...  \n",
       "4    Chapter 219: Gson\\nGson is a Java library that...  \n",
       "..                                                 ...  \n",
       "479  Chapter 148: GameCenter Leaderboards\\nSection ...  \n",
       "480  iOS\\nDeveloper\\niOS Developer\\nNotes for Profe...  \n",
       "481  Chapter 42: UIActivityViewController\\nParamete...  \n",
       "482  Chapter 25: UITextField Delegate\\nSection 25.1...  \n",
       "483  Chapter 166: AirPrint tutorial in iOS\\nSection...  \n",
       "\n",
       "[484 rows x 5 columns]"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_curriculum_dataframe = pd.merge(curriculum_dataframe,english_filtered_content,left_on=\"content_ids\", right_on=\"id\").drop([\"id\", \"language\", \"kind\", \"license\", \"copyright_holder\"],axis=1)\n",
    "new_curriculum_dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "a659e965",
   "metadata": {},
   "outputs": [],
   "source": [
    "def remove_stopwords(text):\n",
    "    '''\n",
    "    Checks language of text then removes stopwords from that language if supported.\n",
    "    '''\n",
    "    lang_code = text[0:2]\n",
    "    for word in stopwords.words(\"english\"):\n",
    "        text = text.replace(' ' + word + ' ', ' ')\n",
    "    return text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "61a8cf6f",
   "metadata": {},
   "outputs": [],
   "source": [
    "#### Apply remove_stopwords() to our data\n",
    "new_curriculum_dataframe[\"new_title\"] = new_curriculum_dataframe.title.apply(remove_stopwords)\n",
    "new_curriculum_dataframe[\"new_text\"] = new_curriculum_dataframe.text.apply(remove_stopwords)\n",
    "new_curriculum_dataframe[\"new_description\"] = new_curriculum_dataframe.description.apply(remove_stopwords)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "7da5f719",
   "metadata": {},
   "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>content_ids</th>\n",
       "      <th>true_values</th>\n",
       "      <th>title</th>\n",
       "      <th>description</th>\n",
       "      <th>text</th>\n",
       "      <th>new_title</th>\n",
       "      <th>new_text</th>\n",
       "      <th>new_description</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>c_0286ded417a2</td>\n",
       "      <td>[0]</td>\n",
       "      <td>Chapter 142: RenderScript</td>\n",
       "      <td>Chapter 142 of the book on Android.</td>\n",
       "      <td>Chapter 142: RenderScript\\nRenderScript is a s...</td>\n",
       "      <td>Chapter 142: RenderScript</td>\n",
       "      <td>Chapter 142: RenderScript\\nRenderScript script...</td>\n",
       "      <td>Chapter 142 book Android.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>c_02a5e75e01f0</td>\n",
       "      <td>[0]</td>\n",
       "      <td>Chapter 161: TransitionDrawable</td>\n",
       "      <td>Chapter 161 of the book on Android.</td>\n",
       "      <td>Chapter 161: TransitionDrawable\\nSection 161.1...</td>\n",
       "      <td>Chapter 161: TransitionDrawable</td>\n",
       "      <td>Chapter 161: TransitionDrawable\\nSection 161.1...</td>\n",
       "      <td>Chapter 161 book Android.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>c_02adc62b7731</td>\n",
       "      <td>[0]</td>\n",
       "      <td>Chapter 31: Supporting Screens With Dierent R...</td>\n",
       "      <td>Chapter 31 of the book on Android.</td>\n",
       "      <td>Chapter 31: Supporting Screens With\\nDierent ...</td>\n",
       "      <td>Chapter 31: Supporting Screens With Dierent R...</td>\n",
       "      <td>Chapter 31: Supporting Screens With\\nDierent ...</td>\n",
       "      <td>Chapter 31 book Android.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>c_04213823e40c</td>\n",
       "      <td>[0]</td>\n",
       "      <td>Chapter 93: RoboGuice</td>\n",
       "      <td>Chapter 93 of the book on Android.</td>\n",
       "      <td>Chapter 93: RoboGuice\\nSection 93.1: Simple ex...</td>\n",
       "      <td>Chapter 93: RoboGuice</td>\n",
       "      <td>Chapter 93: RoboGuice\\nSection 93.1: Simple ex...</td>\n",
       "      <td>Chapter 93 book Android.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>c_044e55056282</td>\n",
       "      <td>[0]</td>\n",
       "      <td>Chapter 219: Gson</td>\n",
       "      <td>Chapter 219 of the book on Android.</td>\n",
       "      <td>Chapter 219: Gson\\nGson is a Java library that...</td>\n",
       "      <td>Chapter 219: Gson</td>\n",
       "      <td>Chapter 219: Gson\\nGson Java library used conv...</td>\n",
       "      <td>Chapter 219 book Android.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      content_ids true_values  \\\n",
       "0  c_0286ded417a2         [0]   \n",
       "1  c_02a5e75e01f0         [0]   \n",
       "2  c_02adc62b7731         [0]   \n",
       "3  c_04213823e40c         [0]   \n",
       "4  c_044e55056282         [0]   \n",
       "\n",
       "                                               title  \\\n",
       "0                          Chapter 142: RenderScript   \n",
       "1                    Chapter 161: TransitionDrawable   \n",
       "2  Chapter 31: Supporting Screens With Dierent R...   \n",
       "3                              Chapter 93: RoboGuice   \n",
       "4                                  Chapter 219: Gson   \n",
       "\n",
       "                           description  \\\n",
       "0  Chapter 142 of the book on Android.   \n",
       "1  Chapter 161 of the book on Android.   \n",
       "2   Chapter 31 of the book on Android.   \n",
       "3   Chapter 93 of the book on Android.   \n",
       "4  Chapter 219 of the book on Android.   \n",
       "\n",
       "                                                text  \\\n",
       "0  Chapter 142: RenderScript\\nRenderScript is a s...   \n",
       "1  Chapter 161: TransitionDrawable\\nSection 161.1...   \n",
       "2  Chapter 31: Supporting Screens With\\nDierent ...   \n",
       "3  Chapter 93: RoboGuice\\nSection 93.1: Simple ex...   \n",
       "4  Chapter 219: Gson\\nGson is a Java library that...   \n",
       "\n",
       "                                           new_title  \\\n",
       "0                          Chapter 142: RenderScript   \n",
       "1                    Chapter 161: TransitionDrawable   \n",
       "2  Chapter 31: Supporting Screens With Dierent R...   \n",
       "3                              Chapter 93: RoboGuice   \n",
       "4                                  Chapter 219: Gson   \n",
       "\n",
       "                                            new_text  \\\n",
       "0  Chapter 142: RenderScript\\nRenderScript script...   \n",
       "1  Chapter 161: TransitionDrawable\\nSection 161.1...   \n",
       "2  Chapter 31: Supporting Screens With\\nDierent ...   \n",
       "3  Chapter 93: RoboGuice\\nSection 93.1: Simple ex...   \n",
       "4  Chapter 219: Gson\\nGson Java library used conv...   \n",
       "\n",
       "             new_description  \n",
       "0  Chapter 142 book Android.  \n",
       "1  Chapter 161 book Android.  \n",
       "2   Chapter 31 book Android.  \n",
       "3   Chapter 93 book Android.  \n",
       "4  Chapter 219 book Android.  "
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_curriculum_dataframe.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "54b6d03a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def preprocess_text(text: str) -> str:\n",
    "    \"\"\"This utility function sanitizes a string by:\n",
    "    - removing links\n",
    "    - removing special characters\n",
    "    - removing numbers\n",
    "    - removing stopwords\n",
    "    - transforming in lowercase\n",
    "    - removing excessive whitespaces\n",
    "    Args:\n",
    "        text (str): the input text you want to clean\n",
    "        remove_stopwords (bool): whether or not to remove stopwords\n",
    "    Returns:\n",
    "        str: the cleaned text\n",
    "    \"\"\"\n",
    "\n",
    "    # remove links\n",
    "    text = re.sub(r\"http\\S+\", \"\", text)\n",
    "    # remove special chars and numbers\n",
    "    text = re.sub(\"[^A-Za-z]+\", \" \", text)\n",
    "    # remove\n",
    "    text = re.sub(\":\",\" \", text)\n",
    "    #\n",
    "    text = re.sub(\"\\n\",\" \", text)\n",
    "    # return text in lower case and stripped of whitespaces\n",
    "    text = text.lower().strip()\n",
    "    return text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "eb660735",
   "metadata": {},
   "outputs": [],
   "source": [
    "new_curriculum_dataframe['text'] = new_curriculum_dataframe['new_text'].apply(lambda x: preprocess_text(x))\n",
    "new_curriculum_dataframe['title'] = new_curriculum_dataframe['new_title'].apply(lambda x: preprocess_text(x))\n",
    "new_curriculum_dataframe['description'] = new_curriculum_dataframe['new_description'].apply(lambda x: preprocess_text(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "c2d1aa95",
   "metadata": {},
   "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>content_ids</th>\n",
       "      <th>true_values</th>\n",
       "      <th>title</th>\n",
       "      <th>description</th>\n",
       "      <th>text</th>\n",
       "      <th>new_title</th>\n",
       "      <th>new_text</th>\n",
       "      <th>new_description</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>c_0286ded417a2</td>\n",
       "      <td>[0]</td>\n",
       "      <td>chapter renderscript</td>\n",
       "      <td>chapter book android</td>\n",
       "      <td>chapter renderscript renderscript scripting la...</td>\n",
       "      <td>Chapter 142: RenderScript</td>\n",
       "      <td>Chapter 142: RenderScript\\nRenderScript script...</td>\n",
       "      <td>Chapter 142 book Android.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>c_02a5e75e01f0</td>\n",
       "      <td>[0]</td>\n",
       "      <td>chapter transitiondrawable</td>\n",
       "      <td>chapter book android</td>\n",
       "      <td>chapter transitiondrawable section animate vie...</td>\n",
       "      <td>Chapter 161: TransitionDrawable</td>\n",
       "      <td>Chapter 161: TransitionDrawable\\nSection 161.1...</td>\n",
       "      <td>Chapter 161 book Android.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>c_02adc62b7731</td>\n",
       "      <td>[0]</td>\n",
       "      <td>chapter supporting screens with di erent resol...</td>\n",
       "      <td>chapter book android</td>\n",
       "      <td>chapter supporting screens with di erent resol...</td>\n",
       "      <td>Chapter 31: Supporting Screens With Dierent R...</td>\n",
       "      <td>Chapter 31: Supporting Screens With\\nDierent ...</td>\n",
       "      <td>Chapter 31 book Android.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>c_04213823e40c</td>\n",
       "      <td>[0]</td>\n",
       "      <td>chapter roboguice</td>\n",
       "      <td>chapter book android</td>\n",
       "      <td>chapter roboguice section simple example robog...</td>\n",
       "      <td>Chapter 93: RoboGuice</td>\n",
       "      <td>Chapter 93: RoboGuice\\nSection 93.1: Simple ex...</td>\n",
       "      <td>Chapter 93 book Android.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>c_044e55056282</td>\n",
       "      <td>[0]</td>\n",
       "      <td>chapter gson</td>\n",
       "      <td>chapter book android</td>\n",
       "      <td>chapter gson gson java library used convert ja...</td>\n",
       "      <td>Chapter 219: Gson</td>\n",
       "      <td>Chapter 219: Gson\\nGson Java library used conv...</td>\n",
       "      <td>Chapter 219 book Android.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      content_ids true_values  \\\n",
       "0  c_0286ded417a2         [0]   \n",
       "1  c_02a5e75e01f0         [0]   \n",
       "2  c_02adc62b7731         [0]   \n",
       "3  c_04213823e40c         [0]   \n",
       "4  c_044e55056282         [0]   \n",
       "\n",
       "                                               title           description  \\\n",
       "0                               chapter renderscript  chapter book android   \n",
       "1                         chapter transitiondrawable  chapter book android   \n",
       "2  chapter supporting screens with di erent resol...  chapter book android   \n",
       "3                                  chapter roboguice  chapter book android   \n",
       "4                                       chapter gson  chapter book android   \n",
       "\n",
       "                                                text  \\\n",
       "0  chapter renderscript renderscript scripting la...   \n",
       "1  chapter transitiondrawable section animate vie...   \n",
       "2  chapter supporting screens with di erent resol...   \n",
       "3  chapter roboguice section simple example robog...   \n",
       "4  chapter gson gson java library used convert ja...   \n",
       "\n",
       "                                           new_title  \\\n",
       "0                          Chapter 142: RenderScript   \n",
       "1                    Chapter 161: TransitionDrawable   \n",
       "2  Chapter 31: Supporting Screens With Dierent R...   \n",
       "3                              Chapter 93: RoboGuice   \n",
       "4                                  Chapter 219: Gson   \n",
       "\n",
       "                                            new_text  \\\n",
       "0  Chapter 142: RenderScript\\nRenderScript script...   \n",
       "1  Chapter 161: TransitionDrawable\\nSection 161.1...   \n",
       "2  Chapter 31: Supporting Screens With\\nDierent ...   \n",
       "3  Chapter 93: RoboGuice\\nSection 93.1: Simple ex...   \n",
       "4  Chapter 219: Gson\\nGson Java library used conv...   \n",
       "\n",
       "             new_description  \n",
       "0  Chapter 142 book Android.  \n",
       "1  Chapter 161 book Android.  \n",
       "2   Chapter 31 book Android.  \n",
       "3   Chapter 93 book Android.  \n",
       "4  Chapter 219 book Android.  "
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_curriculum_dataframe.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "1647572c",
   "metadata": {},
   "outputs": [],
   "source": [
    "new_curriculum_dataframe[\"corpus\"] = new_curriculum_dataframe[\"title\"] + \" \" + new_curriculum_dataframe[\"description\"] + \" \" + new_curriculum_dataframe[\"text\"] \n",
    "new_curriculum_dataframe[\"corpus\"] = new_curriculum_dataframe[\"corpus\"].str.replace(\"chapter\",\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "4bcdee67",
   "metadata": {},
   "outputs": [],
   "source": [
    "# initialize the vectorizer\n",
    "vectorizer = TfidfVectorizer(sublinear_tf=True, min_df=10, max_df=0.95)\n",
    "# fit_transform applies TF-IDF to clean texts - we save the array of vectors in X\n",
    "X = vectorizer.fit_transform(new_curriculum_dataframe[\"text\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "d5b56356",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "\n",
    "# initialize kmeans with 3 centroids\n",
    "kmeans = KMeans(n_clusters=2, random_state=42)\n",
    "# fit the model\n",
    "kmeans.fit(X)\n",
    "# store cluster labels in a variable\n",
    "clusters = kmeans.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "d87798fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "\n",
    "# initialize PCA with 2 components\n",
    "pca = PCA(n_components=2, random_state=42)\n",
    "# pass our X to the pca and store the reduced vectors into pca_vecs\n",
    "pca_vecs = pca.fit_transform(X.toarray())\n",
    "# save our two dimensions into x0 and x1\n",
    "x0 = pca_vecs[:, 0]\n",
    "x1 = pca_vecs[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "2767fae7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# assign clusters and pca vectors to our dataframe \n",
    "new_curriculum_dataframe['cluster'] = clusters\n",
    "new_curriculum_dataframe['x0'] = x0\n",
    "new_curriculum_dataframe['x1'] = x1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "154bee43",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Cluster 0\n",
      "id,this,layout,override,void,string,int,new,public,android\n",
      "\n",
      "Cluster 1\n",
      "the,view,nil,objective,func,swift,self,let,developer,ios\n"
     ]
    }
   ],
   "source": [
    "def get_top_keywords(n_terms):\n",
    "    \"\"\"This function returns the keywords for each centroid of the KMeans\"\"\"\n",
    "    df = pd.DataFrame(X.todense()).groupby(clusters).mean() # groups the TF-IDF vector by cluster\n",
    "    terms = vectorizer.get_feature_names_out() # access tf-idf terms\n",
    "    for i,r in df.iterrows():\n",
    "        print('\\nCluster {}'.format(i))\n",
    "        print(','.join([terms[t] for t in np.argsort(r)[-n_terms:]])) # for each row of the dataframe, find the n terms that have the highest tf idf score\n",
    "            \n",
    "get_top_keywords(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "70ce2303",
   "metadata": {},
   "outputs": [],
   "source": [
    "# map clusters to appropriate labels \n",
    "cluster_map = {0: \"Topic 1\", 1: \"Topic 2\", 2: \"Topic 3\"}\n",
    "# apply mapping\n",
    "new_curriculum_dataframe['cluster'] = new_curriculum_dataframe['cluster'].map(cluster_map)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "id": "b791d8b6-ee7d-4af1-ba56-8b2a17487129",
   "metadata": {},
   "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>content_ids</th>\n",
       "      <th>true_values</th>\n",
       "      <th>title</th>\n",
       "      <th>description</th>\n",
       "      <th>text</th>\n",
       "      <th>new_title</th>\n",
       "      <th>new_text</th>\n",
       "      <th>new_description</th>\n",
       "      <th>corpus</th>\n",
       "      <th>cluster</th>\n",
       "      <th>x0</th>\n",
       "      <th>x1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>c_0286ded417a2</td>\n",
       "      <td>[0]</td>\n",
       "      <td>chapter renderscript</td>\n",
       "      <td>chapter book android</td>\n",
       "      <td>chapter renderscript renderscript scripting la...</td>\n",
       "      <td>Chapter 142: RenderScript</td>\n",
       "      <td>Chapter 142: RenderScript\\nRenderScript script...</td>\n",
       "      <td>Chapter 142 book Android.</td>\n",
       "      <td>renderscript  book android  renderscript rend...</td>\n",
       "      <td>Topic 1</td>\n",
       "      <td>-0.117258</td>\n",
       "      <td>-0.006920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>c_02a5e75e01f0</td>\n",
       "      <td>[0]</td>\n",
       "      <td>chapter transitiondrawable</td>\n",
       "      <td>chapter book android</td>\n",
       "      <td>chapter transitiondrawable section animate vie...</td>\n",
       "      <td>Chapter 161: TransitionDrawable</td>\n",
       "      <td>Chapter 161: TransitionDrawable\\nSection 161.1...</td>\n",
       "      <td>Chapter 161 book Android.</td>\n",
       "      <td>transitiondrawable  book android  transitiond...</td>\n",
       "      <td>Topic 1</td>\n",
       "      <td>-0.246524</td>\n",
       "      <td>0.334416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>c_02adc62b7731</td>\n",
       "      <td>[0]</td>\n",
       "      <td>chapter supporting screens with di erent resol...</td>\n",
       "      <td>chapter book android</td>\n",
       "      <td>chapter supporting screens with di erent resol...</td>\n",
       "      <td>Chapter 31: Supporting Screens With Dierent R...</td>\n",
       "      <td>Chapter 31: Supporting Screens With\\nDierent ...</td>\n",
       "      <td>Chapter 31 book Android.</td>\n",
       "      <td>supporting screens with di erent resolutions ...</td>\n",
       "      <td>Topic 1</td>\n",
       "      <td>-0.113476</td>\n",
       "      <td>0.149865</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>c_04213823e40c</td>\n",
       "      <td>[0]</td>\n",
       "      <td>chapter roboguice</td>\n",
       "      <td>chapter book android</td>\n",
       "      <td>chapter roboguice section simple example robog...</td>\n",
       "      <td>Chapter 93: RoboGuice</td>\n",
       "      <td>Chapter 93: RoboGuice\\nSection 93.1: Simple ex...</td>\n",
       "      <td>Chapter 93 book Android.</td>\n",
       "      <td>roboguice  book android  roboguice section si...</td>\n",
       "      <td>Topic 1</td>\n",
       "      <td>-0.225783</td>\n",
       "      <td>0.065606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>c_044e55056282</td>\n",
       "      <td>[0]</td>\n",
       "      <td>chapter gson</td>\n",
       "      <td>chapter book android</td>\n",
       "      <td>chapter gson gson java library used convert ja...</td>\n",
       "      <td>Chapter 219: Gson</td>\n",
       "      <td>Chapter 219: Gson\\nGson Java library used conv...</td>\n",
       "      <td>Chapter 219 book Android.</td>\n",
       "      <td>gson  book android  gson gson java library us...</td>\n",
       "      <td>Topic 1</td>\n",
       "      <td>-0.138864</td>\n",
       "      <td>-0.129464</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>479</th>\n",
       "      <td>c_faaa92210fe1</td>\n",
       "      <td>[1]</td>\n",
       "      <td>chapter gamecenter leaderboards</td>\n",
       "      <td>chapter book ios developer</td>\n",
       "      <td>chapter gamecenter leaderboards section gamece...</td>\n",
       "      <td>Chapter 148: GameCenter Leaderboards</td>\n",
       "      <td>Chapter 148: GameCenter Leaderboards\\nSection ...</td>\n",
       "      <td>Chapter 148 book iOS Developer.</td>\n",
       "      <td>gamecenter leaderboards  book ios developer  ...</td>\n",
       "      <td>Topic 2</td>\n",
       "      <td>0.205980</td>\n",
       "      <td>-0.043670</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>480</th>\n",
       "      <td>c_faaea6c6ad4c</td>\n",
       "      <td>[1]</td>\n",
       "      <td>content list</td>\n",
       "      <td>section content list book ios developer</td>\n",
       "      <td>ios developer ios developer notes professional...</td>\n",
       "      <td>Content list</td>\n",
       "      <td>iOS\\nDeveloper\\niOS Developer\\nNotes Professio...</td>\n",
       "      <td>Section \"Content list\" book iOS Developer.</td>\n",
       "      <td>content list section content list book ios dev...</td>\n",
       "      <td>Topic 2</td>\n",
       "      <td>0.153380</td>\n",
       "      <td>0.107826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>481</th>\n",
       "      <td>c_fae0b7c7a87a</td>\n",
       "      <td>[1]</td>\n",
       "      <td>chapter uiactivityviewcontroller</td>\n",
       "      <td>chapter book ios developer</td>\n",
       "      <td>chapter uiactivityviewcontroller parameter nam...</td>\n",
       "      <td>Chapter 42: UIActivityViewController</td>\n",
       "      <td>Chapter 42: UIActivityViewController\\nParamete...</td>\n",
       "      <td>Chapter 42 book iOS Developer.</td>\n",
       "      <td>uiactivityviewcontroller  book ios developer ...</td>\n",
       "      <td>Topic 2</td>\n",
       "      <td>0.277720</td>\n",
       "      <td>-0.043786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>482</th>\n",
       "      <td>c_fd75a88c2bb4</td>\n",
       "      <td>[1]</td>\n",
       "      <td>chapter uitextfield delegate</td>\n",
       "      <td>chapter book ios developer</td>\n",
       "      <td>chapter uitextfield delegate section actions u...</td>\n",
       "      <td>Chapter 25: UITextField Delegate</td>\n",
       "      <td>Chapter 25: UITextField Delegate\\nSection 25.1...</td>\n",
       "      <td>Chapter 25 book iOS Developer.</td>\n",
       "      <td>uitextfield delegate  book ios developer  uit...</td>\n",
       "      <td>Topic 2</td>\n",
       "      <td>0.185265</td>\n",
       "      <td>-0.039178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>483</th>\n",
       "      <td>c_ff73c6a38f19</td>\n",
       "      <td>[1]</td>\n",
       "      <td>chapter airprint tutorial ios</td>\n",
       "      <td>chapter book ios developer</td>\n",
       "      <td>chapter airprint tutorial ios section airprint...</td>\n",
       "      <td>Chapter 166: AirPrint tutorial iOS</td>\n",
       "      <td>Chapter 166: AirPrint tutorial iOS\\nSection 16...</td>\n",
       "      <td>Chapter 166 book iOS Developer.</td>\n",
       "      <td>airprint tutorial ios  book ios developer  ai...</td>\n",
       "      <td>Topic 2</td>\n",
       "      <td>0.235294</td>\n",
       "      <td>0.112837</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>484 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        content_ids true_values  \\\n",
       "0    c_0286ded417a2         [0]   \n",
       "1    c_02a5e75e01f0         [0]   \n",
       "2    c_02adc62b7731         [0]   \n",
       "3    c_04213823e40c         [0]   \n",
       "4    c_044e55056282         [0]   \n",
       "..              ...         ...   \n",
       "479  c_faaa92210fe1         [1]   \n",
       "480  c_faaea6c6ad4c         [1]   \n",
       "481  c_fae0b7c7a87a         [1]   \n",
       "482  c_fd75a88c2bb4         [1]   \n",
       "483  c_ff73c6a38f19         [1]   \n",
       "\n",
       "                                                 title  \\\n",
       "0                                 chapter renderscript   \n",
       "1                           chapter transitiondrawable   \n",
       "2    chapter supporting screens with di erent resol...   \n",
       "3                                    chapter roboguice   \n",
       "4                                         chapter gson   \n",
       "..                                                 ...   \n",
       "479                    chapter gamecenter leaderboards   \n",
       "480                                       content list   \n",
       "481                   chapter uiactivityviewcontroller   \n",
       "482                       chapter uitextfield delegate   \n",
       "483                      chapter airprint tutorial ios   \n",
       "\n",
       "                                 description  \\\n",
       "0                       chapter book android   \n",
       "1                       chapter book android   \n",
       "2                       chapter book android   \n",
       "3                       chapter book android   \n",
       "4                       chapter book android   \n",
       "..                                       ...   \n",
       "479               chapter book ios developer   \n",
       "480  section content list book ios developer   \n",
       "481               chapter book ios developer   \n",
       "482               chapter book ios developer   \n",
       "483               chapter book ios developer   \n",
       "\n",
       "                                                  text  \\\n",
       "0    chapter renderscript renderscript scripting la...   \n",
       "1    chapter transitiondrawable section animate vie...   \n",
       "2    chapter supporting screens with di erent resol...   \n",
       "3    chapter roboguice section simple example robog...   \n",
       "4    chapter gson gson java library used convert ja...   \n",
       "..                                                 ...   \n",
       "479  chapter gamecenter leaderboards section gamece...   \n",
       "480  ios developer ios developer notes professional...   \n",
       "481  chapter uiactivityviewcontroller parameter nam...   \n",
       "482  chapter uitextfield delegate section actions u...   \n",
       "483  chapter airprint tutorial ios section airprint...   \n",
       "\n",
       "                                             new_title  \\\n",
       "0                            Chapter 142: RenderScript   \n",
       "1                      Chapter 161: TransitionDrawable   \n",
       "2    Chapter 31: Supporting Screens With Dierent R...   \n",
       "3                                Chapter 93: RoboGuice   \n",
       "4                                    Chapter 219: Gson   \n",
       "..                                                 ...   \n",
       "479               Chapter 148: GameCenter Leaderboards   \n",
       "480                                       Content list   \n",
       "481               Chapter 42: UIActivityViewController   \n",
       "482                   Chapter 25: UITextField Delegate   \n",
       "483                 Chapter 166: AirPrint tutorial iOS   \n",
       "\n",
       "                                              new_text  \\\n",
       "0    Chapter 142: RenderScript\\nRenderScript script...   \n",
       "1    Chapter 161: TransitionDrawable\\nSection 161.1...   \n",
       "2    Chapter 31: Supporting Screens With\\nDierent ...   \n",
       "3    Chapter 93: RoboGuice\\nSection 93.1: Simple ex...   \n",
       "4    Chapter 219: Gson\\nGson Java library used conv...   \n",
       "..                                                 ...   \n",
       "479  Chapter 148: GameCenter Leaderboards\\nSection ...   \n",
       "480  iOS\\nDeveloper\\niOS Developer\\nNotes Professio...   \n",
       "481  Chapter 42: UIActivityViewController\\nParamete...   \n",
       "482  Chapter 25: UITextField Delegate\\nSection 25.1...   \n",
       "483  Chapter 166: AirPrint tutorial iOS\\nSection 16...   \n",
       "\n",
       "                                new_description  \\\n",
       "0                     Chapter 142 book Android.   \n",
       "1                     Chapter 161 book Android.   \n",
       "2                      Chapter 31 book Android.   \n",
       "3                      Chapter 93 book Android.   \n",
       "4                     Chapter 219 book Android.   \n",
       "..                                          ...   \n",
       "479             Chapter 148 book iOS Developer.   \n",
       "480  Section \"Content list\" book iOS Developer.   \n",
       "481              Chapter 42 book iOS Developer.   \n",
       "482              Chapter 25 book iOS Developer.   \n",
       "483             Chapter 166 book iOS Developer.   \n",
       "\n",
       "                                                corpus  cluster        x0  \\\n",
       "0     renderscript  book android  renderscript rend...  Topic 1 -0.117258   \n",
       "1     transitiondrawable  book android  transitiond...  Topic 1 -0.246524   \n",
       "2     supporting screens with di erent resolutions ...  Topic 1 -0.113476   \n",
       "3     roboguice  book android  roboguice section si...  Topic 1 -0.225783   \n",
       "4     gson  book android  gson gson java library us...  Topic 1 -0.138864   \n",
       "..                                                 ...      ...       ...   \n",
       "479   gamecenter leaderboards  book ios developer  ...  Topic 2  0.205980   \n",
       "480  content list section content list book ios dev...  Topic 2  0.153380   \n",
       "481   uiactivityviewcontroller  book ios developer ...  Topic 2  0.277720   \n",
       "482   uitextfield delegate  book ios developer  uit...  Topic 2  0.185265   \n",
       "483   airprint tutorial ios  book ios developer  ai...  Topic 2  0.235294   \n",
       "\n",
       "           x1  \n",
       "0   -0.006920  \n",
       "1    0.334416  \n",
       "2    0.149865  \n",
       "3    0.065606  \n",
       "4   -0.129464  \n",
       "..        ...  \n",
       "479 -0.043670  \n",
       "480  0.107826  \n",
       "481 -0.043786  \n",
       "482 -0.039178  \n",
       "483  0.112837  \n",
       "\n",
       "[484 rows x 12 columns]"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_curriculum_dataframe"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "93c7c41b-ed4a-4e50-b42c-319e2b9f6206",
   "metadata": {},
   "source": [
    "### Plotting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "id": "57f950e9-de67-4d90-bce0-95078d731b5d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "265\n",
      "214\n"
     ]
    }
   ],
   "source": [
    "count_one = 0\n",
    "count_two = 0\n",
    "for i in range(484):\n",
    "    # print(i)\n",
    "    if (new_curriculum_dataframe.loc[i, \"true_values\"] == [0]) and (new_curriculum_dataframe.loc[i, \"cluster\"] == \"Topic 1\"):\n",
    "        count_one += 1\n",
    "    if (new_curriculum_dataframe.loc[i, \"true_values\"] == [1]) and (new_curriculum_dataframe.loc[i, \"cluster\"] == \"Topic 2\"):\n",
    "        count_two += 1\n",
    "\n",
    "print(count_one)\n",
    "print(count_two)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "id": "a09ad7bc-db25-42dd-b92c-159753988c60",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "list"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(new_curriculum_dataframe.loc[0, \"true_values\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "id": "2c58c762",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# set image size\n",
    "plt.figure(figsize=(12, 7))\n",
    "# set a title\n",
    "plt.title(\"TF-IDF + KMeans clustering\", fontdict={\"fontsize\": 18})\n",
    "# set axes names\n",
    "plt.xlabel(\"X0\", fontdict={\"fontsize\": 16})\n",
    "plt.ylabel(\"X1\", fontdict={\"fontsize\": 16})\n",
    "# create scatter plot with seaborn, where hue is the class used to group the data\n",
    "sns.scatterplot(data=new_curriculum_dataframe, x='x0', y='x1', hue='cluster', palette=\"viridis\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a22dca21",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}