notebooks/covid-19-tweets-classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Gghij_-L1jKO"
   },
   "source": [
    "# Import Dependencies"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Install Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 212,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "eMWoqXYnOHPA",
    "outputId": "1153b442-dbbe-4d20-f28d-9491cab7f6e0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: tweepy in /usr/local/lib/python3.7/dist-packages (3.10.0)\n",
      "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.7/dist-packages (from tweepy) (1.3.0)\n",
      "Requirement already satisfied: requests[socks]>=2.11.1 in /usr/local/lib/python3.7/dist-packages (from tweepy) (2.23.0)\n",
      "Requirement already satisfied: six>=1.10.0 in /usr/local/lib/python3.7/dist-packages (from tweepy) (1.15.0)\n",
      "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.7/dist-packages (from requests-oauthlib>=0.7.0->tweepy) (3.1.0)\n",
      "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests[socks]>=2.11.1->tweepy) (2.10)\n",
      "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests[socks]>=2.11.1->tweepy) (1.24.3)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests[socks]>=2.11.1->tweepy) (2020.12.5)\n",
      "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests[socks]>=2.11.1->tweepy) (3.0.4)\n",
      "Requirement already satisfied: PySocks!=1.5.7,>=1.5.6; extra == \"socks\" in /usr/local/lib/python3.7/dist-packages (from requests[socks]>=2.11.1->tweepy) (1.7.1)\n"
     ]
    }
   ],
   "source": [
    "!pip install tweepy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "GuqNPrsD58j7",
    "outputId": "cc90a0a8-373c-45f3-e9aa-1ebb99e033c5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: Sastrawi in /usr/local/lib/python3.7/dist-packages (1.0.1)\n"
     ]
    }
   ],
   "source": [
    "!pip install Sastrawi"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "bjkFGdukOFZd"
   },
   "outputs": [],
   "source": [
    "import tweepy\n",
    "import re"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import API"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "hRoXl4z915LJ"
   },
   "outputs": [],
   "source": [
    "path = \"../\"\n",
    "api_path = \"../api/\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "QsDTZ7sHEILa"
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append(api_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "vG6SGBUKNvxb"
   },
   "outputs": [],
   "source": [
    "import tweepy_api\n",
    "\n",
    "api_key = tweepy_api.api_key()\n",
    "api_secret_key = tweepy_api.api_secret_key()\n",
    "bearer_token = tweepy_api.bearer_token()\n",
    "\n",
    "access_token = tweepy_api.access_token()\n",
    "access_token_secret = tweepy_api.access_token_secret()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "metadata": {
    "id": "GotBFev0OwrZ"
   },
   "outputs": [],
   "source": [
    "auth = tweepy.OAuthHandler(api_key, api_secret_key)\n",
    "auth.set_access_token(access_token, access_token_secret)\n",
    "api = tweepy.API(auth)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2q2Ha_0ppiW1"
   },
   "source": [
    "# Retrieve Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "HIwnWPh-plD0"
   },
   "source": [
    "## Vaccine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "id": "ok6ewP5cnMQh"
   },
   "outputs": [],
   "source": [
    "vaccine_tweets = []"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "BYOHKMI2psBk"
   },
   "source": [
    "### Astrazeneca"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "yNvsrTXceqQp",
    "outputId": "a816cdfa-e289-414d-bb26-eb44c91f8a9c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Menlu_RI 2021-06-10 16:08:53 :\n",
      " Ketiga jenis vaksin yang digunakan Indonesia, Sinovac, AstraZeneca dan Sinopharm telah memperoleh Emergency Use Listing (EUL) dari WHO.\n",
      "------------------------------------------\n",
      "Menlu_RI 2021-06-10 16:08:53 :\n",
      " Alhamdullilah pada malam ini, Indonesia kembali menerima vaksin AstraZeneca melalui jalur multilateral, Covax Facil… https://t.co/uoiAL92MPV\n",
      "------------------------------------------\n",
      "kumparan 2021-06-10 16:08:34 :\n",
      " Menteri Luar Negeri Retno Marsudi mengatakan, vaksin AstraZeneca ini tiba dari skema COVAX. Sebelumnya, pada 5 Juni… https://t.co/DWkhhhj3ha\n",
      "------------------------------------------\n",
      "kompascom 2021-06-10 16:07:03 :\n",
      " Dengan penambahan 1,5 juta dosis vaksin AstraZeneca tersebut, total dosis vaksin yang telah diterima Indonesia seba… https://t.co/tjcH4xGBfx\n",
      "------------------------------------------\n",
      "AlbertSolo2 2021-06-10 16:06:31 :\n",
      " Indonesia Kedatangan 1,5 Juta Dosis Vaksin AstraZeneca https://t.co/cxSlZqh34b\n",
      "------------------------------------------\n",
      "CNNIndonesia 2021-06-10 16:04:14 :\n",
      " Indonesia Kedatangan 1,5 Juta Dosis Vaksin AstraZeneca https://t.co/15wWXUZ9Vw\n",
      "------------------------------------------\n",
      "AyoAjaOrangnya 2021-06-10 15:59:30 :\n",
      " Alhamdulillah. Saya sudah di VAKSIN! #AstraZeneca sebagai bentuk ikhtiar pencegahan penyakit COVID-19. Terima kasih… https://t.co/pa4db8sbfS\n",
      "------------------------------------------\n",
      "EvaMalca 2021-06-10 15:58:37 :\n",
      " @afrkml Aku kira itu list akibat vaksin astrazeneca ☺️\n",
      "------------------------------------------\n",
      "setkabgoid 2021-06-10 15:57:30 :\n",
      " Hari ini, sebanyak 1,5 juta dosis vaksin COVID-19 produksi AstraZeneca kembali tiba di Tanah Air. \n",
      "\n",
      "Selain itu, bes… https://t.co/yHM7WpxYxZ\n",
      "------------------------------------------\n",
      "agitamr 2021-06-10 15:38:47 :\n",
      " nih yang nanya efek samping vaksin astrazeneca :\n",
      "1) Kalo malem panas dingin mengigil\n",
      "2) Badan linu sampe ke tulang… https://t.co/3SWRZEyJay\n",
      "------------------------------------------\n",
      "lensaRTV 2021-06-10 15:38:00 :\n",
      " Indonesia kembali kedatangan vaksin Covid-19 tahap ke-15 yang tiba malam ini di Bandara Soekarno-Hatta. Kali ini ju… https://t.co/XLCIjvZksM\n",
      "------------------------------------------\n",
      "InvestorID 2021-06-10 15:31:28 :\n",
      " Malam Ini, Indonesia Terima 1,5 Juta Dosis Vaksin AstraZeneca https://t.co/O3Vo67WteZ https://t.co/bCS95LMoZE\n",
      "------------------------------------------\n",
      "cnbcindonesia 2021-06-10 15:31:10 :\n",
      " RI Kedatangan 1,5 Juta Dosis Vaksin Covid-19 AstraZeneca https://t.co/mlNdfzQiEZ\n",
      "------------------------------------------\n",
      "Beritasatu 2021-06-10 15:20:18 :\n",
      " Malam Ini, Indonesia Terima 1,5 Juta Dosis Vaksin Astrazeneca https://t.co/eOX5ac7X01 https://t.co/9cnGsU0rab\n",
      "------------------------------------------\n",
      "inibukantessa 2021-06-10 15:10:31 :\n",
      " @dmanda17 Kita abis vaksin astrazeneca berarti jadu keturunan keluarga kerajaan Inggris dong\n",
      "------------------------------------------\n",
      "mohzamzam_ 2021-06-10 15:07:08 :\n",
      " jadi gue dapet vaksin #AstraZeneca di RSUP fatmawati, untuk dosis pertama itu gue disuntik hari ini which is tangga… https://t.co/myXxaluo1Z\n",
      "------------------------------------------\n",
      "mohzamzam_ 2021-06-10 15:03:49 :\n",
      " mau sharing tentang vaksin #AstraZeneca ah untuk beberapa hari/minggu/bulan kedepan. twitnya gue pin ya !!\n",
      "------------------------------------------\n",
      "opsiin 2021-06-10 14:51:22 :\n",
      " Syarat Penerima Vaksin AstraZeneca, Perlu Banget Dicatat! – Detik- https://t.co/XwSncsMQ2h https://t.co/6q5ERYLqKH\n",
      "------------------------------------------\n",
      "opsiin 2021-06-10 14:51:11 :\n",
      " Syarat Penerima Vaksin AstraZeneca, Perlu Banget Dicatat! – Detik https://t.co/mt3LQ3YnDp\n",
      "------------------------------------------\n",
      "syarifahay_ 2021-06-10 14:45:00 :\n",
      " Astrazeneca / sinovac jd abis vaksin maem https://t.co/sKFBupra4U\n",
      "------------------------------------------\n",
      "opsiin 2021-06-10 14:35:48 :\n",
      " DKI Kebut Suntik Vaksin AstraZeneca Cegah Kedaluwarsa Juni – CNN Indonesia- https://t.co/S6Mv5QCQlY https://t.co/GRW965SBzO\n",
      "------------------------------------------\n",
      "opsiin 2021-06-10 14:35:40 :\n",
      " DKI Kebut Suntik Vaksin AstraZeneca Cegah Kedaluwarsa Juni – CNN Indonesia https://t.co/mAY3ARLm7L\n",
      "------------------------------------------\n",
      "ftmhabdllh 2021-06-10 14:31:11 :\n",
      " Feedback dari kami bertiga lepas ambil vaksin yang berlainan.\n",
      "\n",
      "Mak (pfizer) - tak ada apa apa.\n",
      "Abg majid (sinovac)… https://t.co/cMVBfefswI\n",
      "------------------------------------------\n",
      "opsiin 2021-06-10 14:30:50 :\n",
      " Mobil Puskesmas Angkut Vaksin AstraZeneca Terguling di Blitar – detikNews- https://t.co/jcsfpASe9D https://t.co/oj5TmzYrOu\n",
      "------------------------------------------\n",
      "opsiin 2021-06-10 14:30:41 :\n",
      " Mobil Puskesmas Angkut Vaksin AstraZeneca Terguling di Blitar – detikNews https://t.co/h6lMYO1edf\n",
      "------------------------------------------\n",
      "opsiin 2021-06-10 14:29:07 :\n",
      " Kemenkes Sediakan 400 Ribu Vaksin AstraZeneca untuk DKI – Republika- https://t.co/O3W8gaLfPs https://t.co/UPKkSFA5Y0\n",
      "------------------------------------------\n",
      "opsiin 2021-06-10 14:28:51 :\n",
      " Kemenkes Sediakan 400 Ribu Vaksin AstraZeneca untuk DKI – Republika https://t.co/pFTVEAaAHN\n",
      "------------------------------------------\n",
      "giewhydi 2021-06-10 14:21:00 :\n",
      " Daftar vaksin yang telah disetujui WHO.\n",
      "\n",
      "Isndonesia telah menggunakan 3 jenis vaksin, yakni Sinovac, AstraZeneca, d… https://t.co/LggOflKW9b\n",
      "------------------------------------------\n",
      "KATADATAcoid 2021-06-10 14:16:17 :\n",
      " Sebanyak 400 ribu stok vaksin Covid-19 merek AstraZeneca yang berada di Jakarta akan kedaluwarsa pada Juni 2021. Ke… https://t.co/QrGQD4JabF\n",
      "------------------------------------------\n",
      "FIMELAdotcom 2021-06-10 14:16:05 :\n",
      " Cerita Sandiaga Uno Jalani Vaksin Covid-19 AstraZeneca Dosis Pertama: Sandiaga Uno Jalani Vaksin Covid-19 AstraZene… https://t.co/eZItg8U50U\n",
      "------------------------------------------\n",
      "andradjaafara 2021-06-10 14:09:52 :\n",
      " @afriandiniputri ini semoga membantu din https://t.co/ew2kKXL6iS\n",
      "------------------------------------------\n",
      "WilsonLayman 2021-06-10 14:07:21 :\n",
      " @setkabgoid @jokowi Apakah vaksin astrazeneca akan diimpor kembali untuk suntikan kedua atau akan digunakan vaksin… https://t.co/hE65XY7F1f\n",
      "------------------------------------------\n",
      "detikHealth 2021-06-10 14:06:32 :\n",
      " Vaksin Sinovac dan Astrazeneca saat ini jadi vaksin COVID-19 yang paling banyak digunakan di Indonesia. Ini beda be… https://t.co/6mvHvl8iKX\n",
      "------------------------------------------\n",
      "Ahdorable_Trap 2021-06-10 14:03:45 :\n",
      " Aku belum vaksin, \n",
      "ditawarin vaksin  Sinovac atau vaksin Astrazeneca \n",
      "\n",
      "Masih ragu mau vaksin tapi takut jg kena covid 😶\n",
      "------------------------------------------\n",
      "DoyanMusik4 2021-06-10 14:00:13 :\n",
      " Syarat Penerima Vaksin AstraZeneca, Perlu Banget Dicatat! https://t.co/GhMibVxAS8\n",
      "------------------------------------------\n",
      "detikcom 2021-06-10 13:55:40 :\n",
      " Vaksinasi Corona tahap ketiga diperluas sasarannya, boleh untuk masyarakat umum 18 tahun ke atas. Adapun syarat pen… https://t.co/34BhQ5QNV1\n",
      "------------------------------------------\n",
      "detikHealth 2021-06-10 13:54:54 :\n",
      " Masih takut menggunakan vaksin AstraZeneca? Berikut ini beberapa fakta terkait keamanannya:\n",
      "\n",
      "https://t.co/IJPXPaa7Of\n",
      "------------------------------------------\n",
      "datukhensem 2021-06-10 13:54:39 :\n",
      " Hari ini 3046 orang telah diberikan dos vaksin AstraZeneca di PPVAZ BCCK Kuching. \n",
      "\n",
      "Mantap orang Kuching.\n",
      "------------------------------------------\n",
      "afriandiniputri 2021-06-10 13:52:30 :\n",
      " Studi Terbaru Efek Samping Vaksin AstraZeneca, Ini Hasilnya https://t.co/sjhTTy6Znh\n",
      "------------------------------------------\n",
      "detikHealth 2021-06-10 13:49:11 :\n",
      " Pelaksanaan vaksinasi tahap III di DKI ini akan menggunakan vaksin AstraZeneca.\n",
      "\n",
      "https://t.co/yHr3V3G3ZU\n",
      "------------------------------------------\n",
      "detikcom 2021-06-10 13:37:38 :\n",
      " Vaksin Sinovac dan Astrazeneca saat ini jadi vaksin COVID-19 yang paling banyak digunakan di Indonesia. Ini beda be… https://t.co/5DDNaqTEwJ\n",
      "------------------------------------------\n",
      "indoseputarinfo 2021-06-10 13:22:10 :\n",
      " Vaksin-vaksin itu ada yang dipake buat vaksinasi gratis, kayak Sinovac dan Astrazeneca. Ada juga yang dipake buat v… https://t.co/3Ry9tlvj0o\n",
      "------------------------------------------\n",
      "42\n"
     ]
    }
   ],
   "source": [
    "# Retrieve Class \"Vaccine\"\n",
    "search_tweets = api.search(q=\"vaksin astrazeneca\", count=100)\n",
    "count = 0\n",
    "for tweet in search_tweets:\n",
    "  tweet_list = []\n",
    "  if (tweet.lang == 'in' and (re.search(r'^RT @', tweet.text) == None)) :\n",
    "    tweet_list.extend([tweet.user.screen_name, tweet.text, tweet.created_at, 'vaccine'])\n",
    "    vaccine_tweets.append(tweet_list)\n",
    "    print(tweet.user.screen_name, tweet.created_at, \":\\n\", tweet.text)\n",
    "    print(\"------------------------------------------\")\n",
    "    count += 1\n",
    "print(count)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "IsZ-Q1uKpvFT"
   },
   "source": [
    "### Sinovac"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "FWbFNZEgoZnX",
    "outputId": "6d634c24-1f37-4191-dccc-4ade334361cf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "32\n"
     ]
    }
   ],
   "source": [
    "# Retrieve Class \"Vaccine\"\n",
    "search_tweets = api.search(q=\"vaksin sinovac\", count=100)\n",
    "count = 0\n",
    "for tweet in search_tweets:\n",
    "  tweet_list = []\n",
    "  if (tweet.lang == 'in' and (re.search(r'^RT @', tweet.text) == None)) :\n",
    "    tweet_list.extend([tweet.user.screen_name, tweet.text, tweet.created_at, 'vaccine'])\n",
    "    vaccine_tweets.append(tweet_list)\n",
    "    # print(tweet.user.screen_name, tweet.created_at, \":\\n\", tweet.text)\n",
    "    # print(\"------------------------------------------\")\n",
    "    count += 1\n",
    "print(count)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Vd3oXJHXpwl2"
   },
   "source": [
    "### Sinopharm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "UY_PwBKQolEV",
    "outputId": "b09cc8f9-337a-4344-d26c-aa857a732c2c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41\n"
     ]
    }
   ],
   "source": [
    "# Retrieve Class \"Vaccine\"\n",
    "search_tweets = api.search(q=\"vaksin sinopharm\", count=100)\n",
    "count = 0\n",
    "for tweet in search_tweets:\n",
    "  tweet_list = []\n",
    "  if (tweet.lang == 'in' and (re.search(r'^RT @', tweet.text) == None)) :\n",
    "    tweet_list.extend([tweet.user.screen_name, tweet.text, tweet.created_at, 'vaccine'])\n",
    "    vaccine_tweets.append(tweet_list)\n",
    "    # print(tweet.user.screen_name, tweet.created_at, \":\\n\", tweet.text)\n",
    "    # print(\"------------------------------------------\")\n",
    "    count += 1\n",
    "print(count)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "wtH5g2Lhp0-1"
   },
   "source": [
    "## Pencegahan atau Pengobatan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "id": "SlXtOuL2sOsR"
   },
   "outputs": [],
   "source": [
    "pencegahanpengobatan_tweets = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "jrWAXAevp5Zs",
    "outputId": "2bdb7abc-a161-41f1-9a4b-881177e871f6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "54\n"
     ]
    }
   ],
   "source": [
    "# Retrieve Class \"Pencegahan atau Pengobatan\" with keyword \"pencegahan covid-19\"\n",
    "search_tweets = api.search(q=\"pencegahan covid-19\", count=100)\n",
    "count = 0\n",
    "for tweet in search_tweets:\n",
    "  tweet_list = []\n",
    "  if (tweet.lang == 'in' and (re.search(r'^RT @', tweet.text) == None)) :\n",
    "    tweet_list.extend([tweet.user.screen_name, tweet.text, tweet.created_at, 'pencegahan atau pengobatan'])\n",
    "    pencegahanpengobatan_tweets.append(tweet_list)\n",
    "    # print(tweet.user.screen_name, tweet.created_at, \":\\n\", tweet.text)\n",
    "    # print(\"------------------------------------------\")\n",
    "    count += 1\n",
    "print(count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "IlveKvMdqcYF",
    "outputId": "43001a37-aaef-488a-b049-98cca308efa9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "26\n"
     ]
    }
   ],
   "source": [
    "# Retrieve Class \"Pencegahan atau Pengobatan\" with keyword \"pengobatan covid-19\"\n",
    "search_tweets = api.search(q=\"pengobatan covid-19\", count=100)\n",
    "count = 0\n",
    "for tweet in search_tweets:\n",
    "  tweet_list = []\n",
    "  if (tweet.lang == 'in' and (re.search(r'^RT @', tweet.text) == None)) :\n",
    "    tweet_list.extend([tweet.user.screen_name, tweet.text, tweet.created_at, 'pencegahan atau pengobatan'])\n",
    "    pencegahanpengobatan_tweets.append(tweet_list)\n",
    "    # print(tweet.user.screen_name, tweet.created_at, \":\\n\", tweet.text)\n",
    "    # print(\"------------------------------------------\")\n",
    "    count += 1\n",
    "print(count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "LmVbEFvgsan0",
    "outputId": "b66a0cda-6469-46ae-9330-c3d88687bf36"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "72\n"
     ]
    }
   ],
   "source": [
    "# Retrieve Class \"Pencegahan atau Pengobatan\" with keyword \"pencegahan corona\"\n",
    "search_tweets = api.search(q=\"pencegahan corona\", count=100)\n",
    "count = 0\n",
    "for tweet in search_tweets:\n",
    "  tweet_list = []\n",
    "  if (tweet.lang == 'in' and (re.search(r'^RT @', tweet.text) == None)) :\n",
    "    tweet_list.extend([tweet.user.screen_name, tweet.text, tweet.created_at, 'pencegahan atau pengobatan'])\n",
    "    pencegahanpengobatan_tweets.append(tweet_list)\n",
    "    # print(tweet.user.screen_name, tweet.created_at, \":\\n\", tweet.text)\n",
    "    # print(\"------------------------------------------\")\n",
    "    count += 1\n",
    "print(count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Zwh_hHapszNP",
    "outputId": "320e70bd-84cb-4096-9b1f-9d54d0e26561"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n"
     ]
    }
   ],
   "source": [
    "# Retrieve Class \"Pencegahan atau Pengobatan\" with keyword \"pengobatan corona\"\n",
    "search_tweets = api.search(q=\"pengobatan corona\", count=100)\n",
    "count = 0\n",
    "for tweet in search_tweets:\n",
    "  tweet_list = []\n",
    "  if (tweet.lang == 'in' and (re.search(r'^RT @', tweet.text) == None)) :\n",
    "    tweet_list.extend([tweet.user.screen_name, tweet.text, tweet.created_at, 'pencegahan atau pengobatan'])\n",
    "    pencegahanpengobatan_tweets.append(tweet_list)\n",
    "    # print(tweet.user.screen_name, tweet.created_at, \":\\n\", tweet.text)\n",
    "    # print(\"------------------------------------------\")\n",
    "    count += 1\n",
    "print(count)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "TOu5dgHxsoNu"
   },
   "source": [
    "## Perkembangan Covid-19"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "id": "c9cluOp8sqs-"
   },
   "outputs": [],
   "source": [
    "perkembangan_tweets = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "O-tDfQF7s8dJ",
    "outputId": "5740f9a5-9631-4092-f143-61428fdec4ae"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "42\n"
     ]
    }
   ],
   "source": [
    "# Retrieve Class \"Perkembangan Covid-19\" with keyword \"perkembangan covid-19\"\n",
    "search_tweets = api.search(q=\"perkembangan covid-19\", count=100)\n",
    "count = 0\n",
    "for tweet in search_tweets:\n",
    "  tweet_list = []\n",
    "  if (tweet.lang == 'in' and (re.search(r'^RT @', tweet.text) == None)) :\n",
    "    tweet_list.extend([tweet.user.screen_name, tweet.text, tweet.created_at, 'perkembangan covid-19'])\n",
    "    perkembangan_tweets.append(tweet_list)\n",
    "    # print(tweet.user.screen_name, tweet.created_at, \":\\n\", tweet.text)\n",
    "    # print(\"------------------------------------------\")\n",
    "    count += 1\n",
    "print(count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "uEp2ZC_BtNxp",
    "outputId": "2fab53c4-13a6-49b3-cdea-b76ca1d8340d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18\n"
     ]
    }
   ],
   "source": [
    "# Retrieve Class \"Perkembangan Covid-19\" with keyword \"covid-19 di Indonesia\"\n",
    "search_tweets = api.search(q=\"covid-19 di Indonesia\", count=100)\n",
    "count = 0\n",
    "for tweet in search_tweets:\n",
    "  tweet_list = []\n",
    "  if (tweet.lang == 'in' and (re.search(r'^RT @', tweet.text) == None)) :\n",
    "    tweet_list.extend([tweet.user.screen_name, tweet.text, tweet.created_at, 'perkembangan covid-19'])\n",
    "    perkembangan_tweets.append(tweet_list)\n",
    "    # print(tweet.user.screen_name, tweet.created_at, \":\\n\", tweet.text)\n",
    "    # print(\"------------------------------------------\")\n",
    "    count += 1\n",
    "print(count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ey-SHgamtQwW",
    "outputId": "e64c63de-18a6-4224-a89c-816ff5a24668"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "87\n"
     ]
    }
   ],
   "source": [
    "# Retrieve Class \"Perkembangan Covid-19\" with keyword \"covid-19 berkembang\"\n",
    "search_tweets = api.search(q=\"covid-19 berkembang\", count=100)\n",
    "count = 0\n",
    "for tweet in search_tweets:\n",
    "  tweet_list = []\n",
    "  if (tweet.lang == 'in' and (re.search(r'^RT @', tweet.text) == None)) :\n",
    "    tweet_list.extend([tweet.user.screen_name, tweet.text, tweet.created_at, 'perkembangan covid-19'])\n",
    "    perkembangan_tweets.append(tweet_list)\n",
    "    # print(tweet.user.screen_name, tweet.created_at, \":\\n\", tweet.text)\n",
    "    # print(\"------------------------------------------\")\n",
    "    count += 1\n",
    "print(count)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "x3I3yWFXp-Du"
   },
   "source": [
    "# Save Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4DqWP0yxo2Bd",
    "outputId": "921bedd9-3185-42b0-bb14-4517119effb3"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "115"
      ]
     },
     "execution_count": 75,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(vaccine_tweets) #Panjang Kelas Vaccine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "_PLC5l2Fqf2z",
    "outputId": "f93c1ca0-e6f5-4248-e49b-c0d9413a884b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "155"
      ]
     },
     "execution_count": 83,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(pencegahanpengobatan_tweets) #Panjang Kelas Pencegahan & Pengobatan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "SVU6H8oiqg4G",
    "outputId": "bc25953e-ec92-43f6-e669-60d853d25c9c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "147"
      ]
     },
     "execution_count": 87,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(perkembangan_tweets) #Panjang Kelas Perkembangan Covid-19"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ceGN13sGqzIL",
    "outputId": "6235ef85-9719-40ea-af0e-049f4f5a9a5f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "417"
      ]
     },
     "execution_count": 94,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "covid19_tweets = []\n",
    "covid19_tweets = vaccine_tweets + pencegahanpengobatan_tweets + perkembangan_tweets\n",
    "len(covid19_tweets) # Total Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "id": "bmdgGRxuo5D3",
    "outputId": "d6ac66a3-3761-4a84-ccfb-32677b27810f"
   },
   "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>username</th>\n",
       "      <th>tweet</th>\n",
       "      <th>date</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Menlu_RI</td>\n",
       "      <td>Ketiga jenis vaksin yang digunakan Indonesia, ...</td>\n",
       "      <td>2021-06-10 16:08:53</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Menlu_RI</td>\n",
       "      <td>Alhamdullilah pada malam ini, Indonesia kembal...</td>\n",
       "      <td>2021-06-10 16:08:53</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>kumparan</td>\n",
       "      <td>Menteri Luar Negeri Retno Marsudi mengatakan, ...</td>\n",
       "      <td>2021-06-10 16:08:34</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>kompascom</td>\n",
       "      <td>Dengan penambahan 1,5 juta dosis vaksin AstraZ...</td>\n",
       "      <td>2021-06-10 16:07:03</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AlbertSolo2</td>\n",
       "      <td>Indonesia Kedatangan 1,5 Juta Dosis Vaksin Ast...</td>\n",
       "      <td>2021-06-10 16:06:31</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      username  ...    class\n",
       "0     Menlu_RI  ...  vaccine\n",
       "1     Menlu_RI  ...  vaccine\n",
       "2     kumparan  ...  vaccine\n",
       "3    kompascom  ...  vaccine\n",
       "4  AlbertSolo2  ...  vaccine\n",
       "\n",
       "[5 rows x 4 columns]"
      ]
     },
     "execution_count": 96,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.DataFrame(covid19_tweets, columns = ['username', 'tweet', 'date', 'class'])\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {
    "id": "ozs0fs3Gu0bG"
   },
   "outputs": [],
   "source": [
    "df.to_csv('../data/covid19_tweets.csv') # Save to covid19_tweets.csv"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "lnkM4mDeyszd"
   },
   "source": [
    "# Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "metadata": {
    "id": "fK2YaSRKy2ms"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('../data/covid19_tweets.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 221,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "id": "Eidtoxzs2udO",
    "outputId": "7bebaffe-2cd0-43e2-8257-514fe6af31d2"
   },
   "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>Unnamed: 0</th>\n",
       "      <th>username</th>\n",
       "      <th>tweet</th>\n",
       "      <th>date</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>Menlu_RI</td>\n",
       "      <td>Ketiga jenis vaksin yang digunakan Indonesia, ...</td>\n",
       "      <td>2021-06-10 16:08:53</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Menlu_RI</td>\n",
       "      <td>Alhamdullilah pada malam ini, Indonesia kembal...</td>\n",
       "      <td>2021-06-10 16:08:53</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>kumparan</td>\n",
       "      <td>Menteri Luar Negeri Retno Marsudi mengatakan, ...</td>\n",
       "      <td>2021-06-10 16:08:34</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>kompascom</td>\n",
       "      <td>Dengan penambahan 1,5 juta dosis vaksin AstraZ...</td>\n",
       "      <td>2021-06-10 16:07:03</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>AlbertSolo2</td>\n",
       "      <td>Indonesia Kedatangan 1,5 Juta Dosis Vaksin Ast...</td>\n",
       "      <td>2021-06-10 16:06:31</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0     username  ...                 date    class\n",
       "0           0     Menlu_RI  ...  2021-06-10 16:08:53  vaccine\n",
       "1           1     Menlu_RI  ...  2021-06-10 16:08:53  vaccine\n",
       "2           2     kumparan  ...  2021-06-10 16:08:34  vaccine\n",
       "3           3    kompascom  ...  2021-06-10 16:07:03  vaccine\n",
       "4           4  AlbertSolo2  ...  2021-06-10 16:06:31  vaccine\n",
       "\n",
       "[5 rows x 5 columns]"
      ]
     },
     "execution_count": 221,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "JHgluoWT2yhl",
    "outputId": "9bd8fff3-36ca-497c-b374-d6200be1df3f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 417 entries, 0 to 416\n",
      "Data columns (total 5 columns):\n",
      " #   Column      Non-Null Count  Dtype \n",
      "---  ------      --------------  ----- \n",
      " 0   Unnamed: 0  417 non-null    int64 \n",
      " 1   username    417 non-null    object\n",
      " 2   tweet       417 non-null    object\n",
      " 3   date        417 non-null    object\n",
      " 4   class       417 non-null    object\n",
      "dtypes: int64(1), object(4)\n",
      "memory usage: 16.4+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 173
    },
    "id": "4X1KGBVU4YZ_",
    "outputId": "29810302-3ee0-4db3-b578-658500d49252"
   },
   "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>Unnamed: 0</th>\n",
       "      <th>username</th>\n",
       "      <th>tweet</th>\n",
       "      <th>date</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>class</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>pencegahan atau pengobatan</th>\n",
       "      <td>155</td>\n",
       "      <td>155</td>\n",
       "      <td>155</td>\n",
       "      <td>155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>perkembangan covid-19</th>\n",
       "      <td>147</td>\n",
       "      <td>147</td>\n",
       "      <td>147</td>\n",
       "      <td>147</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vaccine</th>\n",
       "      <td>115</td>\n",
       "      <td>115</td>\n",
       "      <td>115</td>\n",
       "      <td>115</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            Unnamed: 0  username  tweet  date\n",
       "class                                                        \n",
       "pencegahan atau pengobatan         155       155    155   155\n",
       "perkembangan covid-19              147       147    147   147\n",
       "vaccine                            115       115    115   115"
      ]
     },
     "execution_count": 223,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('class').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 401
    },
    "id": "uarxuqwNxVmT",
    "outputId": "3a906154-ce12-4cef-d66e-d2de1433c56e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0, 0, 'vaccine'),\n",
       " Text(0, 0, 'pencegahan atau pengobatan'),\n",
       " Text(0, 0, 'perkembangan covid-19')]"
      ]
     },
     "execution_count": 224,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmQAAAFKCAYAAACzX0NnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3debhkVX3u8e8LTTOogEiLyCCgqAE0UVricM0VNYqJinG6eh1wxHkIibPGIWpM1KgYTSQRwZHggHIVB4IDSgRtRBRFIwpII0PLPDcNv/vHWsUpDqfh0HadfbrP9/M85+mqXVW7VlXvqv3u31p7VaoKSZIkDWeDoRsgSZK00BnIJEmSBmYgkyRJGpiBTJIkaWAGMkmSpIEZyCRJkga2aOgG/CG23nrr2mmnnYZuhiRJ0i068cQTf19VS2a6bWKBLMnBwKOB86tqj7HlLwNeAlwHfKWqXt2Xvw54bl/+8qr6+i09x0477cSyZcsm0XxJkqS1KsmZq7ttkhWyQ4B/AT4+1pC9gX2BP66qa5LcsS/fDXgKsDtwZ+C/kty9qq6bYPskSZLmhYmNIauqY4ELpy1+EfCuqrqm3+f8vnxf4LCquqaqTgdOA/aaVNskSZLmk7ke1H934MFJTkjynST368u3A84au9/yvkySJGm9N9eD+hcBWwH3B+4HHJ5kl1uzgiT7A/sD7Ljjjmu9gZIkSXNtritky4EvVPMD4Hpga+BsYIex+23fl91EVR1UVUuraumSJTOeqCBJkrROmetA9kVgb4AkdwcWA78HjgSekmTjJDsDuwI/mOO2SZIkDWKS0158BngIsHWS5cCbgYOBg5OcAqwE9quqAn6W5HDg58Aq4CWeYSlJkhaKtDy0blq6dGk5D5kkSVoXJDmxqpbOdJs/nSRJkjQwA5kkSdLADGSSJEkDW6d/XFzSrfPbt91r6CZoPbPj3/106CZI6wUrZJIkSQMzkEmSJA3MQCZJkjQwA5kkSdLADGSSJEkDM5BJkiQNzEAmSZI0MAOZJEnSwAxkkiRJAzOQSZIkDcxAJkmSNDADmSRJ0sD8cXFJ0nrlQR980NBN0HrmuJcdN/HnsEImSZI0MAOZJEnSwAxkkiRJAzOQSZIkDcxAJkmSNDADmSRJ0sAMZJIkSQObWCBLcnCS85OcMsNtf5OkkmzdryfJgUlOS/KTJPedVLskSZLmm0lWyA4B9pm+MMkOwCOA344tfhSwa//bH/jXCbZLkiRpXplYIKuqY4ELZ7jpfcCrgRpbti/w8WqOB7ZMsu2k2iZJkjSfzOkYsiT7AmdX1cnTbtoOOGvs+vK+TJIkab03Z79lmWQz4PW07so/ZD3707o12XHHHddCyyRJkoY1lxWyuwI7AycnOQPYHvhRkjsBZwM7jN13+77sJqrqoKpaWlVLlyxZMuEmS5IkTd6cBbKq+mlV3bGqdqqqnWjdkvetqnOBI4Fn9rMt7w9cUlXnzFXbJEmShjTJaS8+A3wfuEeS5UmeezN3Pwr4DXAa8O/AiyfVLkmSpPlmYmPIquqpt3D7TmOXC3jJpNoiSZI0nzlTvyRJ0sAMZJIkSQMzkEmSJA3MQCZJkjQwA5kkSdLADGSSJEkDM5BJkiQNzEAmSZI0MAOZJEnSwAxkkiRJAzOQSZIkDcxAJkmSNDADmSRJ0sAMZJIkSQMzkEmSJA3MQCZJkjQwA5kkSdLADGSSJEkDM5BJkiQNzEAmSZI0MAOZJEnSwAxkkiRJAzOQSZIkDcxAJkmSNDADmSRJ0sAmFsiSHJzk/CSnjC17d5JfJPlJkiOSbDl22+uSnJbkl0keOal2SZIkzTeTrJAdAuwzbdnRwB5VdW/gf4DXASTZDXgKsHt/zIeTbDjBtkmSJM0bEwtkVXUscOG0Zd+oqlX96vHA9v3yvsBhVXVNVZ0OnAbsNam2SZIkzSdDjiF7DvDVfnk74Kyx25b3ZTeRZP8ky5IsW7FixYSbKEmSNHmDBLIkbwBWAZ+6tY+tqoOqamlVLV2yZMnab5wkSdIcWzTXT5jkWcCjgYdVVfXFZwM7jN1t+75MkiRpvTenFbIk+wCvBh5bVVeO3XQk8JQkGyfZGdgV+MFctk2SJGkoE6uQJfkM8BBg6yTLgTfTzqrcGDg6CcDxVfXCqvpZksOBn9O6Ml9SVddNqm2SJEnzycQCWVU9dYbFH72Z+78DeMek2iNJkjRfOVO/JEnSwAxkkiRJAzOQSZIkDWzOp72Yr/Z81ceHboLWMye++5lDN0GStI6wQiZJkjQwA5kkSdLADGSSJEkDM5BJkiQNzEAmSZI0MAOZJEnSwAxkkiRJAzOQSZIkDcxAJkmSNDADmSRJ0sAMZJIkSQMzkEmSJA3MQCZJkjQwA5kkSdLADGSSJEkDM5BJkiQNzEAmSZI0MAOZJEnSwAxkkiRJAzOQSZIkDWxigSzJwUnOT3LK2LKtkhyd5Ff939v35UlyYJLTkvwkyX0n1S5JkqT5ZpIVskOAfaYtey1wTFXtChzTrwM8Cti1/+0P/OsE2yVJkjSvTCyQVdWxwIXTFu8LHNovHwo8bmz5x6s5HtgyybaTapskSdJ8MtdjyLapqnP65XOBbfrl7YCzxu63vC+TJEla7w02qL+qCqhb+7gk+ydZlmTZihUrJtAySZKkuTXXgey8UVdk//f8vvxsYIex+23fl91EVR1UVUuraumSJUsm2lhJkqS5MNeB7Ehgv355P+BLY8uf2c+2vD9wyVjXpiRJ0npt0aRWnOQzwEOArZMsB94MvAs4PMlzgTOBJ/e7HwX8BXAacCXw7Em1S5Ikab6ZWCCrqqeu5qaHzXDfAl4yqbZIkiTNZ87UL0mSNDADmSRJ0sAMZJIkSQMzkEmSJA3MQCZJkjQwA5kkSdLADGSSJEkDM5BJkiQNzEAmSZI0MAOZJEnSwAxkkiRJAzOQSZIkDcxAJkmSNDADmSRJ0sBmFciSHDObZZIkSbr1Ft3cjUk2ATYDtk5yeyD9ps2B7SbcNkmSpAXhZgMZ8ALglcCdgROZCmSXAv8ywXZJkiQtGDcbyKrqA8AHkrysqj44R22SJElaUG6pQgZAVX0wyQOBncYfU1Ufn1C7JEmSFoxZBbIknwDuCvwYuK4vLsBAJkmS9AeaVSADlgK7VVVNsjGSJEkL0WznITsFuNMkGyJJkrRQzbZCtjXw8yQ/AK4ZLayqx06kVZIkSQvIbAPZWybZCEmSpIVstmdZfmdtPmmSvwaeRzsx4KfAs4FtgcOAO9DmPHtGVa1cm88rSZI0H832p5MuS3Jp/7s6yXVJLl2TJ0yyHfByYGlV7QFsCDwF+EfgfVV1N+Ai4Llrsn5JkqR1zawCWVXdrqo2r6rNgU2BJwAf/gOedxGwaZJFtJ9mOgd4KPC5fvuhwOP+gPVLkiStM2Z7luUNqvki8Mg1ecKqOht4D/BbWhC7hNZFeXFVrep3W46/lSlJkhaI2U4M+/ixqxvQ5iW7ek2esP9I+b7AzsDFwGeBfW7F4/cH9gfYcccd16QJkiRJ88psz7J8zNjlVcAZtFC1Jh4OnF5VKwCSfAF4ELBlkkW9SrY9cPZMD66qg4CDAJYuXepEtZIkaZ0327Msn70Wn/O3wP2TbAZcBTwMWAZ8C3gi7UzL/YAvrcXnlCRJmrdme5bl9kmOSHJ+//t8ku3X5Amr6gTa4P0f0aa82IBW8XoNcECS02hTX3x0TdYvSZK0rpltl+XHgE8DT+rXn96X/fmaPGlVvRl487TFvwH2WpP1SZIkrctme5blkqr6WFWt6n+HAEsm2C5JkqQFY7aB7IIkT0+yYf97OnDBJBsmSZK0UMw2kD0HeDJwLm3usCcCz5pQmyRJkhaU2Y4hexuwX1VdBJBkK9rkrs+ZVMMkSZIWitlWyO49CmMAVXUhcJ/JNEmSJGlhmW0g26DPsA/cUCGbbXVNkiRJN2O2oeq9wPeTfLZffxLwjsk0SZIkaWGZ7Uz9H0+yDHhoX/T4qvr55JolSZK0cMy627EHMEOYJEnSWjbbMWSSJEmaEAOZJEnSwAxkkiRJAzOQSZIkDcxAJkmSNDADmSRJ0sAMZJIkSQMzkEmSJA3MQCZJkjQwA5kkSdLADGSSJEkDM5BJkiQNzEAmSZI0MAOZJEnSwAxkkiRJAzOQSZIkDWyQQJZkyySfS/KLJKcmeUCSrZIcneRX/d/bD9E2SZKkuTZUhewDwNeq6p7AHwOnAq8FjqmqXYFj+nVJkqT13pwHsiRbAH8GfBSgqlZW1cXAvsCh/W6HAo+b67ZJkiQNYYgK2c7ACuBjSU5K8h9JbgNsU1Xn9PucC2wzQNskSZLm3BCBbBFwX+Bfq+o+wBVM656sqgJqpgcn2T/JsiTLVqxYMfHGSpIkTdoQgWw5sLyqTujXP0cLaOcl2Rag/3v+TA+uqoOqamlVLV2yZMmcNFiSJGmS5jyQVdW5wFlJ7tEXPQz4OXAksF9fth/wpblumyRJ0hAWDfS8LwM+lWQx8Bvg2bRweHiS5wJnAk8eqG2SJElzapBAVlU/BpbOcNPD5rotkiRJQ3OmfkmSpIEZyCRJkgZmIJMkSRqYgUySJGlgBjJJkqSBGcgkSZIGZiCTJEkamIFMkiRpYAYySZKkgRnIJEmSBmYgkyRJGpiBTJIkaWAGMkmSpIEZyCRJkgZmIJMkSRqYgUySJGlgBjJJkqSBGcgkSZIGZiCTJEkamIFMkiRpYAYySZKkgRnIJEmSBmYgkyRJGpiBTJIkaWCDBbIkGyY5KcmX+/Wdk5yQ5LQk/5lk8VBtkyRJmktDVsheAZw6dv0fgfdV1d2Ai4DnDtIqSZKkOTZIIEuyPfCXwH/06wEeCnyu3+VQ4HFDtE2SJGmuDVUhez/wauD6fv0OwMVVtapfXw5sN0TDJEmS5tqcB7IkjwbOr6oT1/Dx+ydZlmTZihUr1nLrJEmS5t4QFbIHAY9NcgZwGK2r8gPAlkkW9ftsD5w904Or6qCqWlpVS5csWTIX7ZUkSZqoOQ9kVfW6qtq+qnYCngJ8s6qeBnwLeGK/237Al+a6bZIkSUOYT/OQvQY4IMlptDFlHx24PZIkSXNi0S3fZXKq6tvAt/vl3wB7DdkeSZKkIcynCpkkSdKCZCCTJEkamIFMkiRpYAYySZKkgRnIJEmSBmYgkyRJGpiBTJIkaWAGMkmSpIEZyCRJkgZmIJMkSRqYgUySJGlgBjJJkqSBGcgkSZIGZiCTJEkamIFMkiRpYAYySZKkgRnIJEmSBmYgkyRJGpiBTJIkaWAGMkmSpIEZyCRJkgZmIJMkSRqYgUySJGlgBjJJkqSBGcgkSZIGNueBLMkOSb6V5OdJfpbkFX35VkmOTvKr/u/t57ptkiRJQxiiQrYK+Juq2g24P/CSJLsBrwWOqapdgWP6dUmSpPXenAeyqjqnqn7UL18GnApsB+wLHNrvdijwuLlumyRJ0hAGHUOWZCfgPsAJwDZVdU6/6Vxgm9U8Zv8ky5IsW7FixZy0U5IkaZIGC2RJbgt8HnhlVV06fltVFVAzPa6qDqqqpVW1dMmSJXPQUkmSpMkaJJAl2YgWxj5VVV/oi89Lsm2/fVvg/CHaJkmSNNeGOMsywEeBU6vqn8duOhLYr1/eD/jSXLdNkiRpCIsGeM4HAc8Afprkx33Z64F3AYcneS5wJvDkAdomSZI05+Y8kFXV94Cs5uaHzWVbJEmS5gNn6pckSRqYgUySJGlgBjJJkqSBGcgkSZIGZiCTJEkamIFMkiRpYAYySZKkgRnIJEmSBmYgkyRJGpiBTJIkaWAGMkmSpIEZyCRJkgZmIJMkSRqYgUySJGlgBjJJkqSBGcgkSZIGZiCTJEkamIFMkiRpYAYySZKkgRnIJEmSBmYgkyRJGpiBTJIkaWAGMkmSpIEZyCRJkgY27wJZkn2S/DLJaUleO3R7JEmSJm1eBbIkGwIfAh4F7AY8Ncluw7ZKkiRpsuZVIAP2Ak6rqt9U1UrgMGDfgdskSZI0UfMtkG0HnDV2fXlfJkmStN5aNHQDbq0k+wP796uXJ/nlkO1ZgLYGfj90I9YFec9+QzdBa87tfLbenKFboDXndj5Lefla287vsrob5lsgOxvYYez69n3ZDarqIOCguWyUpiRZVlVLh26HNElu51oI3M7nl/nWZflDYNckOydZDDwFOHLgNkmSJE3UvKqQVdWqJC8Fvg5sCBxcVT8buFmSJEkTNa8CGUBVHQUcNXQ7tFp2F2shcDvXQuB2Po+kqoZugyRJ0oI238aQSZIkLTjzrstS664ko/OCA1BV14/fVpZjpXXG6PM8+tz26xsA1Rf7eda8lWQr4J607fWnVXX5wE26RVbIdLPGQhZJNkiy4eivL/t6krtA/4Zurh8PY6Pb+v03TuJ2Jw0gzQbjn8Ek90qydHT7aPnowzzt+nX9820Y00SMb4OzuO/9kxye5JNJ/mRs+c60n2F8F/Bq4Hlrv6VrnzvGBWrsi/mGcDWT8aPj/kV83eiv3/7Iqjqzr2eXJA9O8uIkL0qyZX/sJkkOTPJj4HvAY5NYnZX+QNMDVpIdkuyf5H79+o2+41dzwHQv4P+MrXKDJDsleWCSB40+q0n+tH+O35fk5CRvHD1g4i9U67W+zd07yd5VVf36Rv22RyX5Vr98owIBsBFwMXBb4G5jq3wYsElV/RnwMuC+Se4zV69nTblTXKB60FrtUW4v9+4A3A44taouSPLnwJOAewCfqap/S3IM8Mn+9+/ASuC7tB+I3xD4F+AVtJLxy5PcC3gjcA3wVbsypRub/pkY7ypcTeV5/PNzG+B+wJn99vFhA6PbRpNvfxpYAjyfqR3Wx4BlwH8A5wBXAPcFPkj73D8f+FPgi8Bf9x3ot9bOK9f6poemMLbtzvSdX1XX97lHV46uA6Nt92xg8758/HEFHAecStvHbNbXvymwGDij3+/i/u/DgZPW4stb6wxkC0Q/2ljVjz6W0H6+4Y+BPwHuDhwBfKKqrkiyJ/BmYCvah+EtSW4HPBn4Oe1L+6K+6vOAO1bVtX3H8a2q+qckZwN7jT3Xa5I8k3YkcxVwLdzkAyat92baSY2bVpUedRtet5r13BO4K3B1VR0NXErbAT0myZtpO7V3VNVXgZ2AFwKn0wLe26rqDUm+SAtyj2Cq1+QvgU2BvwDeRgtkZwG/raqfJNkY+CmwB2AgWwCmjymcjZvbvsfWuS2wihagduvLt6B1OW7bl2/Yt/eatp5KchmtUrZZX35Vkh8CL0yyF+0Xf3YGzp39qx2GgWwB6FWp/WlB6kfAX9G6KY4DjqZt+C8Fnk2raP018P2q+oexdbwO2LCq3jdt9efQNnhoX9BX9MsraEc7O9BC20er6lVr/cVJA1qTCu9MO6m+rtvQfltwB+C8qvpVX74n8I/AhcCvga2q6gW0StU/0XY0SbJLVX0kyW1pQWlf4I+Ajyb5k6r6WZIDgB2B/wW8FngD7TvgaaOqRB9q8O+9LRfSdnq3B84HruzNvR64nFZB1zpqfPsdG7pyswcKt7C+zYEtgbOr6rokDwIeTatYfZC2rX6Itj/4b9qBwCeAt9Cqta8A/jfwDNq29ryx65tU1ZXc1ErgatoBxKitP0zyJuCttH3Uf/U2zGuOIVsYfk6rSj26X98BuKyqDgN2AV4MPAG4Z5LtaWHqRIAkm/XHXE/foJPcdtS/Tztq3q5fvgS4Q798IbAx7Uv718CemToRYPcku0/gdUprJMni3k1/QyUgUyewrHaM1M3tpPpj75zkj8bHTCZ5VpL/TPL5JA/uy55AG1/598A7gA/0Nm0O/ANtAs93A3vSKwG0ndeHq+pJtGEAr+9tvQRYVlUrgP8GjgcekuQOTB1w3RG4Ksm2tGr1+M7qCcBVVbV3VT2BVn3YklYVrySLqupa2nfKJtPHqWl4fdsbjRFe7f/PtJM2RuODbxTGkmyR5K5JHp7kvqt5vrv0cV7fBg4B7tS3rTfQqrbn0Ia1bETbJp/VH3o9sFNVfZd24D7qeXkiLbSdXlWH0Lbp2yfZPG0g/+P7gcPoNawEbtMrt6PX86WqelRVPYc2vuz4W3rfhmaFbAHoRyo/BfbppeBVwCVJHk87Uv4nWgj7E1pXykp6yBo7IvlZf/wdquqCsdWfyVQgu5rWLQLt6OaOwJ2Bw4DbAyf2HdPZtCMiac6Nuj6mhantgf2AN4+Wj05cmeGxd6qq3/XL96EduR83dp/NgA/TPmd70A58/xN4b9+h7UwLWQHeleRptMry9lV1n76Or9MGJp8K7FhVh/flBwNP7U/1R8DLkyyuql8kuZ4WnC4HLujBaVWSVb0NT6QdiO3X1/Us2rCEM4Atxl5mAaPqxj2AbWif5V/SupC2plU6rqPtYDdhqnKmtagH7HvQJnE/dfR/ekuPm2nbHVvnbWgHztcBd62qY3tYfwPtc3Ab4ICq+mWShwDvo/V+3IEWip5YVb+bttqPAP9WVf851u7nA+eOelrSxiDvTQtsr+/DWfYCvtGD1BXAxv2xmwGXj73ea2jb6J7AM2nb84lMjQ87k/a5vGbsdd6TVsHdkbadL7ul921oBrKF4zhaWHoL7ajku8DjgS9X1Q/SThnevarOSvILYL8k36XtPC4DvkEba/bJJNfRPjwvBH5Hq4RBq8SNuixPBz4A/KLvFD5MO+Lxi1t/kP6FPfq7YQqG2XYfrqbLcHfgOUl+Qtvev0vbEe1A2zm8sYeuewBfo42LhNYtuJj2+Rqt/8rexjtV1f3TBhmfnuSTtDC1J22Q8h2BB9B2ShcDy5Js1KtPP6Ed6JwHnJZks/7ZOYupatbFwG5V9e1+fSPaiTQrgOfQqmGraAdd/9jvs3GS5/Tn3xrYtndlfr+/9mXAAcBSWkXuq7SK3Xm0KsVT+r8AH5lNONDq9V6DYtq2PH4X4AW09//Umd7vfmCwFXBpVa3s29tf0fbve9N6K17dD8z3AD5OCzgnA09Ksl1vw7G0cL4RcGCSvwIuoAWax1TV8iSfBR6e5JN9IP7oM7cr7TNDP0BYmWQX4DtJNq+qS4EfAtv1dnwdeCjwOOALVXVNkqtp1drbAicAj6+qv09yx/76dqiqI4Ejp73+t/T3aHGSvwMeW1XfoQ3TuT+tQveWqjpn1v8xAzGQLRBV9ZsesL4MHFdVf5NkJ+AFSf4Xrf/++iSb0M6+Wgx8hrYD+GpVHZ/kQ7TQtaovv6SqTqBV1qiqL44932inNrp+HR5FawZjAeuGsJTkz4CnVtWLpt9/NJh3/PHV9eu3pVVkz5m+A+vVq0NoQWtT4K1VdQTwGFplYB/gm/3uR9C6268DjkgbIHwhcFGSjftO5HxgyQyViyuB4/v9rkpyAq3b5C600PNjWpXpTbSd4L3684zGaq2kBcFf0w6Etgf+B3gksEWvcnwKeGbvutkN+H+0Lp+N+vN/IcndaBXq3/Yq2n1pZ0AfQ6tgnNrb+1LagdUFVbWSNoxhJt8Z+38wjHVj2/BfAmdWO/HhFgfBr66S1UPVyh5ezqV1DR9A2xY+3ataGwBfoG0bAT6d5EBaxfKjtED9JeBFwNNoQexfaQcXR/XK7IOBLavq3CRX0cZsbQv8Oa0idjnwA6aGN/2Utq/YlHbwHdpn8Yq+/Hd9+4G27T4A+Fy//sDeLoBDaZWuxwLv6cuuZ2r6ijfRQuFP+vOfT9uuRyE2wHX9vT2Jtv3+hnaQ8tv+3h6wuvd9vjKQLSwn0b5QR2dFfYZ2RHIx8Atad8bV/baP9L8b9HLw1+amqVpXzRSw+vKZugpXNwXLYnr3d3/cplV1Rb++J63qcxdaBfagvvzztG61BwN3At6R5NBpz3ctbZzWGbQAdHKSY6tq/yRLq+r5Y+2F1o2zBLg37YzG02g7jiXActoOaxTuLkuyQX/NV9KqUKPxZ6MuopOA21fVp8de22LaeKyiVQLO75fvVFWXJPkA8KG+k/6ffvt2VXVQD5hPo1UB3tl34P+PduB1PXDF+BCDqnorM6iqC6cvm6l6M9sq5LpupjCVqbFYq92Gkzyd1gV3aq90jq9zS1r379XV5m5cDDyXFoQfSutV+CRtPOD9gL+lHRxfRjtQOJ42dOS9SZ7VK5vPpX1/V3/8f1XVyf1A4ah+IH0n4N79YHtj2nguaAccLwbu2NvyNNrBwttoJ4zsQqvUXk3bfn9LC16b0LbZK2hB7frericm+R0tVK2kTY3yp8A/J7mWdgByfH+/TkjyCtqByWjbW0E7meSiqro4yetpn6HltH3TjEMJqupLrEamzmheJyYzNpAtIFV1Ce0IbnT9KtoZLtLNGoWsaQHrJstg9XPcTXvsqMK0CHgILXxtAxzcuxYuBVYlOYw2DuucJE+tqt/TqgTQugkfneT0alM+bATcuaoenPbrEZ8Dvs7UDohq07PsDryXtuO4K3CPJCf259ujqk5Jm+n7WbSzsz5Lq5rt1Csf19OC03Jat+IWtB3dZUwFsPNoc/Z9L8lonMtltDPN3p3kCNr3752ZOrq/iKkB+8f392nTqvpQkqP7Op9CC4dX9tfzfuD94/8nVXXm9Pd+/Hamfv4oq6vQ9HXf5LZ1Yae2Nsz0Oqdv5wBpZ59uTOvyvoQWlhf17ewvaBPuvpR2QtVzaNvH2UneThv39DZatejt/fLbaNvd5rQQ/j1akNkYeH9VnZf2qwoPo4Wgx9AmPr2aVtm6K60r8jdM7d9X0MYb3obWBbkd7cBgc9pnZvQZvKKqDuyva0PaQcfltAOJO/Z1raJ9Tm9D215H78lLaDPif6Ov851V9ekk76JNp7IY+FpV/Xrs/fy/097fi2hFAvqBzfLp7/dMxg4AZwrKM57RPF8ZyKQFaE0rAOOPX13w6juMXWhVo8N68LotcDBtDNaXgacmeSAtpDyS9sV/JfCmJG8Afk+rEjynd68cAeyb5GO0oLU3LcTdl7YjPJpWKTi2dx+e2V/itrQwt0G1MS93p437+peqOjLJt4Gdq+q/k6ykd4v09RZwaLUxkFszNW7sPODJvWK1F21nsGVv8yiQXU3r3nklbWf4TtoYoOuTvK2v/ypaWPx17+Z55uj/oaq+PPZ+7kEbe7Y7bSf5D9XG86RXrUbTFdxiFaDfvtoQttD1/9MtaP/Xm//bAdkAAAjvSURBVNCGd6zqBw5vp1V8ijYmaxmtF2ExrfL62b6aK5PsQztJ5EBaeN8b+NtevTqCFs7eTqtqfa/aON7v0w4SrqmqX41tv1fRzsC9HW3bWwZsmnaG7mi930pyOC2QQdsWRydbjVeyfg48LMkyWlDanFZNPoY2NuxY2vitTWgnmVyb5Bza2ET6c59B+9zeEHiq6sIkr6uq1469l6mq02hV5dW93zNNgnyrgtTqvofWRQYyaQG6lRWAxbQgdTfgG1W1vAeBJ9C6PO4EfJ42+HsPWjfMpbQv9dvRBpe/BTiFNvj2ybQv0C2qjV15T3/czsDf0aZ4OIs2vml0NtfxtJ3TIuDptDEuP6J1v2+adpbWxbQQOHptl/a2wdQO6Z60o/zRKfA7MRW0zgb2T3JUf/4LgROS/Kg/dhS2/rb/vY7WhT+qbsFUxeByWmXigKoaVchGO6BzgK9Mf69Hamoc3Ya9SnUerdL3FVpV5Zx+v9WeDapbr4euo2iB9SLadnEgrRvxBbTt8e20MP3itJ+OOoV2AsSzq+ryJHemVcOuBT5eVcelnc2+F/DAXi29khbEoP1/btIvX0wLPqODgtH2exYt1G9DCzf3oFXNrqVt6yvTzp7/Y9q2MjL6KaFVtM/h3YFX0Q4O/ou2Tf2CNrH32Wnjz+5Gq569jxbqqKqX9fdngxo7m3i60fY4/WBvFgd6q72+0BjIpAVkDSoAB/Xrl9ArUr3CszEtGL2X9uV+u14JewctOBxD2zG9MslXaNWdf6uqi3o35JNpY69+Cby+r/trtCCzC23ndz5tkDu0ndNdaDuhZwH/UVUfT/Ii2mzyG/X7b0ULX6OTSHaZ9hZ8k9Y9c3yS42mhaXRm8IG0cTR70roW30s7e+zXtO6YiwGq6lRa6Lw5l/X3dwvg4lG4Wt1OayY19XuxK3o7NUH9c3A74OvVfsHg4cDz0n6D90G0g4ZH0A4ctqVVl66hHRiMQvEFTE338+m+bBPa5+H9VXXW+HMmuYAWtKCFwB2YOmt99Jk7ua/vxUn+krZNfqHa+LAH0U7aOJ42HuuU/thPMnU27Am07s1RBffvq+q1SR5BG1R/Un/9y7iZqSHGDhRG4yRXd791uttwSAYyaYFYgwrAG2hf8PcCXlhtwPg3aUfRd6UFle/07rYLeti7jhZW9qSFkn+mVZ6uoHXHQNuJXUU7Tf3htBD1f6rq6iTPoHWVXJPkCqYmGr68P/72tO7Jv0o7Xf/BtO+xJbTQdjdaleFa2gSUowHDo3BzeZJ30mYLP7f6iQL9tuMYm76CtnM9nRmMDRYeVamuH/8X+HxVjbqwZhqIvKArAfPYr5mqyp5Eq0Q9gDag/QKmxuud17elR9G2vfGu6h8Bh9POgL0tbVzUI2g/5fN+WrWsaFOKrKJVbUePHZ3BCPCrfvsl/Xk3oE1V8k6mttMPVNV7p7+I6gPde0X2MnoXY9rkx3/Tq3bn0KYiOnms+/tmw1ZftwFrQgxk0gKxBhWALWg7iZ/Qviuuo+2UtqFVpC7jxmM3rqF1mRxWVTc6WSTt7Kv706Zq2I125L8lbXLHlcAb037/9I/680PbAW3bL59PO8JfTJt09Zm9fe9jqhvvI7RT4Vf11zuakPJG41SqnRjw+5neo9F4rFGAmv7YsXW401o/raBVoGDq9xGX0w4Etq4+KD3JHdIm3P0drctytC8dDa7/EG3erTfRDnwOoE0G/HXawdARPQB9gql55b5CGxIwqsS+a6xdM558VWNTj4wGt99cJava2bRv6H83PG6s+9vtdkAGMmlhWdMKwGgcyHm0HcxRtDFjfwZ8LX3yxz4w+XF957A5rcL1CdrEpG9Mcgate+U0WiXs00mOpA1aP4M22Hl0luAB9O7Eqvo6bWc2Mv03VWc0myP+cVayFryLgSf0EylGk/5+mzbdyEv7oPfb0AL9y2jb6gqmDkxOoh1grKg29+PzaNNcXEarQt9ItXkcR6Hoium3j52MssG0x63utybr5u4zE7fx+cNAJi0sf2gF4HLgAVX1/rRZu1+V5H3A/yR5R1X9e690/V9aF+LJtDmEViR5FW0KgLvQBsWfDjd0r9xkLqGaNjfW+JH8tC7DmwwWHluHR/y6NX5GmxroebTxXP9WbXqgXyd5K62r/gLg/Jr61ZEbfiOx2lQNy8eu32js3/h2O5sxWTN0hWs9ZiCTFpY/tALwBaa6ET9PG8uyEvj9WLfHUbQK2g36WJpn0QbNbwIcXVWfHRu7MpqElL6O1VUAVnu7tBZcDlxcVa8cLRhto70r8cTpD5geqGYTsG5pmRamWK2UFo4kT6WdMfgdpioAR/fbtmTmCsBM67nZGdszNTfW6MftQztTbHFf9026Z6Shpf081puq6jFJNqKNSbzRZMh28WlSDGTSApLkMbS5sfYeW3ZL4Wp6BWA0P5YkaS3Z4JbvImk9ch6tW4YkG/WwdZOZssfNcKaWYUyS1jIrZJIkSQOzQiZJkjQwA5kkSdLADGSSJEkDM5BJWpCSvCXJ3w7dDkkCA5kkSdLgDGSSFoQkz0zykyQn9x91Hr/t+Ul+2G/7fJLN+vInJTmlLz+2L9s9yQ+S/Livb9eZnk+Sbg2nvZC03kuyO3AE8MCq+n2SrYCXA5dX1XuS3KGqLuj3fTvtx9U/mOSnwD5VdXaSLavq4iQfBI6vqk8lWQxs2H/vUJLWmBUySQvBQ4HPVtXv4aY/XA7skeS7PYA9Ddi9Lz8OOCTJ84HRz0F9H3h9ktcAdzGMSVobDGSSBIcAL62qewFvpf0AOlX1QuCNtN/9PLFX0j4NPBa4CjgqyUOHabKk9YmBTNJC8E3gSUnuANC7LMfdDjin/6D000YLk9y1qk6oqr8DVgA7JNkF+E1VHQh8Cbj3nLwCSeu1RUM3QJImrap+luQdwHeSXAecBJwxdpc3ASfQQtcJtIAG8O4+aD/AMcDJwGuAZyS5FjgXeOecvAhJ6zUH9UuSJA3MLktJkqSBGcgkSZIGZiCTJEkamIFMkiRpYAYySZKkgRnIJEmSBmYgkyRJGpiBTJIkaWD/H9I+QyvX4m+GAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10,5))\n",
    "chart = sns.countplot(x='class', data=df, ax = ax)\n",
    "chart.set_xticklabels(chart.get_xticklabels(), rotation=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nyFLi6gY4KMQ"
   },
   "source": [
    "## Remove hashtag, @user, link of a tweet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {
    "id": "MjNQrTHD3aXl"
   },
   "outputs": [],
   "source": [
    "new_tweets = []\n",
    "for i in df['tweet'].values :\n",
    "  modified = ' '.join(re.sub(\"(@[A-Za-z0-9]+)|([^0-9A-Za-z \\t])|(\\w+:\\/\\/\\S+)\",\" \",i).split())\n",
    "  new_tweets.append(modified)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 226,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "id": "SA0-pkME45_S",
    "outputId": "16a06b76-3075-4bab-eaa1-e20ac4972d7c"
   },
   "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>Unnamed: 0</th>\n",
       "      <th>username</th>\n",
       "      <th>tweet</th>\n",
       "      <th>date</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>Menlu_RI</td>\n",
       "      <td>Ketiga jenis vaksin yang digunakan Indonesia S...</td>\n",
       "      <td>2021-06-10 16:08:53</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Menlu_RI</td>\n",
       "      <td>Alhamdullilah pada malam ini Indonesia kembali...</td>\n",
       "      <td>2021-06-10 16:08:53</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>kumparan</td>\n",
       "      <td>Menteri Luar Negeri Retno Marsudi mengatakan v...</td>\n",
       "      <td>2021-06-10 16:08:34</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>kompascom</td>\n",
       "      <td>Dengan penambahan 1 5 juta dosis vaksin AstraZ...</td>\n",
       "      <td>2021-06-10 16:07:03</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>AlbertSolo2</td>\n",
       "      <td>Indonesia Kedatangan 1 5 Juta Dosis Vaksin Ast...</td>\n",
       "      <td>2021-06-10 16:06:31</td>\n",
       "      <td>vaccine</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0     username  ...                 date    class\n",
       "0           0     Menlu_RI  ...  2021-06-10 16:08:53  vaccine\n",
       "1           1     Menlu_RI  ...  2021-06-10 16:08:53  vaccine\n",
       "2           2     kumparan  ...  2021-06-10 16:08:34  vaccine\n",
       "3           3    kompascom  ...  2021-06-10 16:07:03  vaccine\n",
       "4           4  AlbertSolo2  ...  2021-06-10 16:06:31  vaccine\n",
       "\n",
       "[5 rows x 5 columns]"
      ]
     },
     "execution_count": 226,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['tweet'] = new_tweets\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "HBLOs-g76vq7"
   },
   "source": [
    "## Train TF-IDF Vectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "YLTRRk7b5mRS",
    "outputId": "7c8e74bc-c0e8-4e10-9b58-b3e0efde4ad4"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['yang', 'untuk', 'pada', 'ke', 'para', 'namun', 'menurut', 'antara', 'dia', 'dua', 'ia', 'seperti', 'jika', 'jika', 'sehingga', 'kembali', 'dan', 'tidak', 'ini', 'karena', 'kepada', 'oleh', 'saat', 'harus', 'sementara', 'setelah', 'belum', 'kami', 'sekitar', 'bagi', 'serta', 'di', 'dari', 'telah', 'sebagai', 'masih', 'hal', 'ketika', 'adalah', 'itu', 'dalam', 'bisa', 'bahwa', 'atau', 'hanya', 'kita', 'dengan', 'akan', 'juga', 'ada', 'mereka', 'sudah', 'saya', 'terhadap', 'secara', 'agar', 'lain', 'anda', 'begitu', 'mengapa', 'kenapa', 'yaitu', 'yakni', 'daripada', 'itulah', 'lagi', 'maka', 'tentang', 'demi', 'dimana', 'kemana', 'pula', 'sambil', 'sebelum', 'sesudah', 'supaya', 'guna', 'kah', 'pun', 'sampai', 'sedangkan', 'selagi', 'sementara', 'tetapi', 'apakah', 'kecuali', 'sebab', 'selain', 'seolah', 'seraya', 'seterusnya', 'tanpa', 'agak', 'boleh', 'dapat', 'dsb', 'dst', 'dll', 'dahulu', 'dulunya', 'anu', 'demikian', 'tapi', 'ingin', 'juga', 'nggak', 'mari', 'nanti', 'melainkan', 'oh', 'ok', 'seharusnya', 'sebetulnya', 'setiap', 'setidaknya', 'sesuatu', 'pasti', 'saja', 'toh', 'ya', 'walau', 'tolong', 'tentu', 'amat', 'apalagi', 'bagaimanapun']\n"
     ]
    }
   ],
   "source": [
    "from Sastrawi.StopWordRemover.StopWordRemoverFactory import StopWordRemoverFactory\n",
    " \n",
    "factory = StopWordRemoverFactory()\n",
    "stopwords = factory.get_stop_words()\n",
    "print(stopwords)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "metadata": {
    "id": "itp4nbAT5MmM"
   },
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "vectorizer = TfidfVectorizer(stop_words = stopwords)\n",
    "features = vectorizer.fit_transform(df['tweet'].values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "CDxymakI7LmX"
   },
   "source": [
    "## Label Encoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "metadata": {
    "id": "HIT74_HQ7M7e"
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "le = LabelEncoder()\n",
    "df['class'] = le.fit_transform(df['class'].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "HWbdfXfy7j20",
    "outputId": "a7c2ee98-a96a-4b4a-a4ee-06fa561a2eda"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['pencegahan atau pengobatan', 'perkembangan covid-19', 'vaccine'],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 230,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "le.classes_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "id": "vec5GPUl7cNN",
    "outputId": "8c967a24-8adc-4dd0-c756-ff3ada36c0fa"
   },
   "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>Unnamed: 0</th>\n",
       "      <th>username</th>\n",
       "      <th>tweet</th>\n",
       "      <th>date</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>Menlu_RI</td>\n",
       "      <td>Ketiga jenis vaksin yang digunakan Indonesia S...</td>\n",
       "      <td>2021-06-10 16:08:53</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Menlu_RI</td>\n",
       "      <td>Alhamdullilah pada malam ini Indonesia kembali...</td>\n",
       "      <td>2021-06-10 16:08:53</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>kumparan</td>\n",
       "      <td>Menteri Luar Negeri Retno Marsudi mengatakan v...</td>\n",
       "      <td>2021-06-10 16:08:34</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>kompascom</td>\n",
       "      <td>Dengan penambahan 1 5 juta dosis vaksin AstraZ...</td>\n",
       "      <td>2021-06-10 16:07:03</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>AlbertSolo2</td>\n",
       "      <td>Indonesia Kedatangan 1 5 Juta Dosis Vaksin Ast...</td>\n",
       "      <td>2021-06-10 16:06:31</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0     username  ...                 date class\n",
       "0           0     Menlu_RI  ...  2021-06-10 16:08:53     2\n",
       "1           1     Menlu_RI  ...  2021-06-10 16:08:53     2\n",
       "2           2     kumparan  ...  2021-06-10 16:08:34     2\n",
       "3           3    kompascom  ...  2021-06-10 16:07:03     2\n",
       "4           4  AlbertSolo2  ...  2021-06-10 16:06:31     2\n",
       "\n",
       "[5 rows x 5 columns]"
      ]
     },
     "execution_count": 231,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "XGJ564LU8AZZ"
   },
   "source": [
    "## Split Data Train & Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "metadata": {
    "id": "s3PRCkJ97sh9"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "x_train, x_test, y_train, y_test = train_test_split(df['tweet'].values, df['class'].values, test_size = 0.2, random_state = 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "L7r5Sb_g8aMr"
   },
   "source": [
    "# Train Naive Bayes Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2WR6kWSP96Cg"
   },
   "source": [
    "## Convert Tweets to Vector"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "metadata": {
    "id": "7t3qb2in8h66"
   },
   "outputs": [],
   "source": [
    "x_train = vectorizer.transform(x_train).toarray()\n",
    "x_test = vectorizer.transform(x_test).toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 234,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "_mKAPGbX9q9k",
    "outputId": "b3043f99-045c-4596-c05f-e83d1c61bb60"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(333, 1615)"
      ]
     },
     "execution_count": 234,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7YEZxltI-DHx"
   },
   "source": [
    "## Train Multinomial Naive Bayes Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "metadata": {
    "id": "kOehWLwd8Obg"
   },
   "outputs": [],
   "source": [
    "from sklearn.naive_bayes import MultinomialNB\n",
    "\n",
    "MNB_model = MultinomialNB()\n",
    "MNB_model.fit(x_train, y_train)\n",
    "y_pred = MNB_model.predict(x_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "jT7pxITs-io4"
   },
   "source": [
    "## Save Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 196,
   "metadata": {
    "id": "OcbZFJYM-jq8"
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "\n",
    "pickle.dump(MNB_model, open(\"../models/MNB_model.sav\", 'wb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DP4usHNS-Hup"
   },
   "source": [
    "# Calculate Performance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "k4GlrI_T_cmY"
   },
   "source": [
    "## Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 569
    },
    "id": "m_8AzdPm-fFg",
    "outputId": "b6505632-8c59-409f-eff7-915b1fa915e1"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(66.47999999999999, 0.5, 'Actual Class')"
      ]
     },
     "execution_count": 236,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "conf_mat = confusion_matrix(y_test, y_pred)\n",
    "ax = plt.subplots(figsize=(9, 9))\n",
    "ax = sns.heatmap(conf_mat, square = True, annot=True, fmt = 'd', cmap=\"YlGnBu\", xticklabels = le.classes_, yticklabels = le.classes_, cbar = False)\n",
    "plt.xlabel(\"Predicted Class\")\n",
    "plt.ylabel(\"Actual Class\")\n",
    "plt.savefig('../results/conf_mat.png', dpi=fig.dpi)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "w8vhr_Ku_e6O"
   },
   "source": [
    "## Accuracy Score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "AHCrlJE78Zl3",
    "outputId": "082d019d-42ea-4ad7-f08e-d8220a17c1a3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nilai akurasi    :  0.9285714285714286\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score, classification_report\n",
    "\n",
    "print(\"Nilai akurasi    : \", accuracy_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Tl0bgabs_i7U"
   },
   "source": [
    "## Classification Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "vdjftpJW-YmF",
    "outputId": "3a44036d-060f-42b9-b8e2-3fbaa03671e1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            precision    recall  f1-score   support\n",
      "\n",
      "pencegahan atau pengobatan       0.90      0.93      0.91        28\n",
      "     perkembangan covid-19       0.97      0.90      0.93        31\n",
      "                   vaccine       0.92      0.96      0.94        25\n",
      "\n",
      "                  accuracy                           0.93        84\n",
      "                 macro avg       0.93      0.93      0.93        84\n",
      "              weighted avg       0.93      0.93      0.93        84\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, y_pred, target_names=le.classes_))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tJ30YLhjJjRO"
   },
   "source": [
    "# Predict Sentence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 250,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "9BWKIi2QJf5d",
    "outputId": "95981c5f-50fc-468f-81db-d476f1bf3a71"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sentence        : Angka pertumbuhan covid-19 di Indonesia semakin tinggi\n",
      "Class Predicted : perkembangan covid-19\n"
     ]
    }
   ],
   "source": [
    "sentence = \"Angka pertumbuhan covid-19 di Indonesia semakin tinggi\"\n",
    "sentence_vector = vectorizer.transform([sentence])\n",
    "print(\"Sentence        :\", sentence)\n",
    "print(\"Class Predicted :\", le.classes_[MNB_model.predict(sentence_vector)[0]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "wcVkhkgWKyOP",
    "outputId": "c5b3883a-340b-4052-e2a2-67b96ca9058b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sentence        : Indonesia Kembali Kedatangan 1.504.800 Vaksin AstraZeneca\n",
      "Class Predicted : vaccine\n"
     ]
    }
   ],
   "source": [
    "sentence = \"Indonesia Kembali Kedatangan 1.504.800 Vaksin AstraZeneca\"\n",
    "sentence_vector = vectorizer.transform([sentence])\n",
    "print(\"Sentence        :\", sentence)\n",
    "print(\"Class Predicted :\", le.classes_[MNB_model.predict(sentence_vector)[0]])"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "authorship_tag": "ABX9TyNApBc52LXR+7by4Ir5whoJ",
   "collapsed_sections": [],
   "include_colab_link": true,
   "name": "pba_tugas klasifikasi tweet covid-19.ipynb",
   "provenance": [],
   "toc_visible": true
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}