augmentation_with_yelp.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- https://ubuntu.com/blog/getting-started-with-cuda-on-ubuntu-on-wsl-2\n",
    "- https://huggingface.co/nlptown/bert-base-multilingual-uncased-sentiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "from googletrans import Translator # Translation library\n",
    "import json # JSON library\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "import matplotlib.pyplot as plt # plotting\n",
    "import numpy as np # linear algebra\n",
    "import seaborn as sns # visualization\n",
    "import matplotlib as mpl\n",
    "\n",
    "\n",
    "# Set image resolution\n",
    "mpl.rcParams['figure.dpi'] = 300\n",
    "sns.set_theme()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "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>Title</th>\n",
       "      <th>Opinion</th>\n",
       "      <th>Polarity</th>\n",
       "      <th>Attraction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Pésimo lugar</td>\n",
       "      <td>Piensen dos veces antes de ir a este hotel, te...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Hotel</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>No vayas a lugar de Eddie</td>\n",
       "      <td>Cuatro de nosotros fuimos recientemente a Eddi...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Restaurant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Mala relación calidad-precio</td>\n",
       "      <td>seguiré corta y simple: limpieza\\n- bad. Tengo...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Hotel</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Minusválido? ¡No te alojes aquí!</td>\n",
       "      <td>Al reservar un hotel con multipropiedad Mayan ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Hotel</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Es una porqueria no pierdan su tiempo</td>\n",
       "      <td>No pierdan su tiempo ni dinero, venimos porque...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Hotel</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                   Title  \\\n",
       "0                           Pésimo lugar   \n",
       "1              No vayas a lugar de Eddie   \n",
       "2           Mala relación calidad-precio   \n",
       "3       Minusválido? ¡No te alojes aquí!   \n",
       "4  Es una porqueria no pierdan su tiempo   \n",
       "\n",
       "                                             Opinion  Polarity  Attraction  \n",
       "0  Piensen dos veces antes de ir a este hotel, te...       1.0       Hotel  \n",
       "1  Cuatro de nosotros fuimos recientemente a Eddi...       1.0  Restaurant  \n",
       "2  seguiré corta y simple: limpieza\\n- bad. Tengo...       1.0       Hotel  \n",
       "3  Al reservar un hotel con multipropiedad Mayan ...       1.0       Hotel  \n",
       "4  No pierdan su tiempo ni dinero, venimos porque...       1.0       Hotel  "
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_excel('data/train_sentiment_analysis.xlsx')\n",
    "data.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of hotels: 16565\n",
      "Number of restaurants: 8450\n",
      "Number of reviews: 30212\n"
     ]
    }
   ],
   "source": [
    "n_hotel = len(data[data['Attraction'] == 'Hotel'])\n",
    "n_rest = len(data[data['Attraction'] == 'Restaurant'])\n",
    "n = len(data)\n",
    "\n",
    "print(f'Number of hotels: {n_hotel}')\n",
    "print(f'Number of restaurants: {n_rest}')\n",
    "print(f'Number of reviews: {n}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Hotels with label 1.0: 259\n",
      "Hotels with label 2.0: 307\n",
      "Hotels with label 3.0: 768\n",
      "Hotels with label 4.0: 2339\n",
      "Hotels with label 5.0: 12892\n",
      "Restaurants with label 1.0: 208\n",
      "Restaurants with label 2.0: 278\n",
      "Restaurants with label 3.0: 667\n",
      "Restaurants with label 4.0: 1943\n",
      "Restaurants with label 5.0: 5354\n",
      "Reviews with label 1.0: 547\n",
      "Reviews with label 2.0: 730\n",
      "Reviews with label 3.0: 2121\n",
      "Reviews with label 4.0: 5878\n",
      "Reviews with label 5.0: 20936\n"
     ]
    }
   ],
   "source": [
    "count_hotel_label = {}\n",
    "for label in data.Polarity.unique():\n",
    "    count_hotel_label[label] = len(data[(data['Attraction'] == 'Hotel') & (data['Polarity'] == label)])\n",
    "    print(f'Hotels with label {label}: {count_hotel_label[label]}')\n",
    "\n",
    "count_rest_label = {}\n",
    "for label in data.Polarity.unique():\n",
    "    count_rest_label[label] = len(data[(data['Attraction'] == 'Restaurant') & (data['Polarity'] == label)])\n",
    "    print(f'Restaurants with label {label}: {count_rest_label[label]}')\n",
    "\n",
    "count_label = {}\n",
    "for label in data.Polarity.unique():\n",
    "    count_label[label] = len(data[data['Polarity'] == label])\n",
    "    print(f'Reviews with label {label}: {count_label[label]}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x900 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(4,3))\n",
    "\n",
    "ax.hist(data['Polarity'], bins=[0.001, 1.001, 2.001, 3.001, 4.001, 5.001], color = '#f98948')\n",
    "ax.set_title('Polarity distribution - Original data.')\n",
    "ax.set_xlabel('Polarity')\n",
    "ax.set_ylabel('Number of reviews')\n",
    "ax.grid(True)\n",
    "ax.set_yticks([0, 10000, 20000], [0, '10k', '20k'])\n",
    "ax.set_xticks([0.5, 1.5, 2.5, 3.5, 4.5], [1, 2, 3, 4, 5])\n",
    "plt.tight_layout()\n",
    "plt.savefig('images/polarity_distribution_original.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generator for getting lines from json file\n",
    "def get_line(path):\n",
    "    with open(path) as a_file:\n",
    "        for line in a_file:\n",
    "            yield line"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [],
   "source": [
    "review_path = 'data/yelp/yelp_academic_dataset_review.json'\n",
    "business_path = 'data/yelp/yelp_academic_dataset_business.json'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'business_id': 'Pns2l4eNsfO8kk83dixA6A',\n",
       " 'name': 'Abby Rappoport, LAC, CMQ',\n",
       " 'address': '1616 Chapala St, Ste 2',\n",
       " 'city': 'Santa Barbara',\n",
       " 'state': 'CA',\n",
       " 'postal_code': '93101',\n",
       " 'latitude': 34.4266787,\n",
       " 'longitude': -119.7111968,\n",
       " 'stars': 5.0,\n",
       " 'review_count': 7,\n",
       " 'is_open': 0,\n",
       " 'attributes': {'ByAppointmentOnly': 'True'},\n",
       " 'categories': 'Doctors, Traditional Chinese Medicine, Naturopathic/Holistic, Acupuncture, Health & Medical, Nutritionists',\n",
       " 'hours': None}"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Open business dataset\n",
    "business_list = []\n",
    "with open(business_path) as a_file:\n",
    "    for line in a_file:\n",
    "        business_list.append(json.loads(line))\n",
    "\n",
    "business_list[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [],
   "source": [
    "restaurant_id = []\n",
    "hotel_id = []\n",
    "\n",
    "for business in business_list:\n",
    "    categories = business['categories']\n",
    "    if categories:\n",
    "        categories = categories.split(',')\n",
    "        if 'Restaurants' in categories:\n",
    "            restaurant_id.append(business['business_id'])\n",
    "        if 'Hotels' in categories:\n",
    "            hotel_id.append(business['business_id'])\n",
    "\n",
    "del business_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "15290 836\n",
      "xM6LoUcnpDpMBzXs_7dXAg\n"
     ]
    }
   ],
   "source": [
    "print(len(restaurant_id), len(hotel_id))\n",
    "print(hotel_id[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'review_id': 'KU_O5udG6zpxOg-VcAEodg',\n",
       " 'user_id': 'mh_-eMZ6K5RLWhZyISBhwA',\n",
       " 'business_id': 'XQfwVwDr-v0ZS3_CbbE5Xw',\n",
       " 'stars': 3.0,\n",
       " 'useful': 0,\n",
       " 'funny': 0,\n",
       " 'cool': 0,\n",
       " 'text': \"If you decide to eat here, just be aware it is going to take about 2 hours from beginning to end. We have tried it multiple times, because I want to like it! I have been to it's other locations in NJ and never had a bad experience. \\n\\nThe food is good, but it takes a very long time to come out. The waitstaff is very young, but usually pleasant. We have just had too many experiences where we spent way too long waiting. We usually opt for another diner or restaurant on the weekends, in order to be done quicker.\",\n",
       " 'date': '2018-07-07 22:09:11'}"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reviews = get_line(review_path) # Generator for getting reviews\n",
    "review = json.loads(next(reviews)) # Review example\n",
    "\n",
    "review"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "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>Attraction</th>\n",
       "      <th>Opinion</th>\n",
       "      <th>Polarity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [Attraction, Opinion, Polarity]\n",
       "Index: []"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yelp_df = pd.DataFrame(columns=['Attraction', 'Opinion', 'Polarity'])\n",
    "yelp_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Added 49900 reviews\n",
      "{1.0: 11519, 2.0: 5647, 3.0: 6398, 4.0: 11689, 5.0: 12892}\n",
      "{1.0: 5354, 2.0: 5354, 3.0: 5354, 4.0: 5354, 5.0: 5354}\n",
      "No more reviews\n"
     ]
    }
   ],
   "source": [
    "# Max number of reviews per label\n",
    "top_hotel = max(count_hotel_label.values()) # Hotels\n",
    "top_rest = max(count_rest_label.values()) # Restaurants\n",
    "top = max(count_label.values()) # All reviews\n",
    "\n",
    "count = 0 # Number of reviews added\n",
    "\n",
    "# While labels invalanced...\n",
    "while (min(count_hotel_label.values()) < top_hotel) or (min(count_rest_label.values()) < top_rest):\n",
    "    try:\n",
    "        review = json.loads(next(reviews)) # Load next review\n",
    "    except StopIteration:\n",
    "        print('No more reviews')\n",
    "        break\n",
    "\n",
    "    id = review['business_id'] # Get business id\n",
    "\n",
    "    # If id hotel or restaurant and label is not top yet...\n",
    "    if (id in hotel_id and count_hotel_label[review['stars']]<top_hotel)\\\n",
    "        or (id in restaurant_id and count_rest_label[review['stars']]<top_rest):\n",
    "\n",
    "        count += 1 # Update count\n",
    "\n",
    "        # Review dictionary to add to dataframe\n",
    "        review_dict = {'Attraction': None,\n",
    "                        'Opinion': review['text'],\n",
    "                        'Polarity': review['stars']}\n",
    "\n",
    "        # Count hotel or restaurant\n",
    "        if review['business_id'] in hotel_id:\n",
    "            review_dict['Attraction'] = 'Hotel'\n",
    "            count_hotel_label[review['stars']] += 1\n",
    "            count_label[review['stars']] += 1\n",
    "        else:\n",
    "            review_dict['Attraction'] = 'Restaurant'\n",
    "            count_rest_label[review['stars']] += 1\n",
    "            count_label[review['stars']] += 1\n",
    "\n",
    "        # Add review to dataframe\n",
    "        yelp_df = pd.concat([yelp_df, pd.DataFrame(review_dict, index=[0])])\n",
    "        if count % 100 == 0:\n",
    "            clear_output(wait=True)\n",
    "            print(f'Added {count} reviews')\n",
    "            print(count_hotel_label)\n",
    "            print(count_rest_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "49919\n"
     ]
    },
    {
     "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>Attraction</th>\n",
       "      <th>Opinion</th>\n",
       "      <th>Polarity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Restaurant</td>\n",
       "      <td>Family diner. Had the buffet. Eclectic assortm...</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Restaurant</td>\n",
       "      <td>The bun makes the Sonoran Dog. It's like a snu...</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Restaurant</td>\n",
       "      <td>Yes, this is the only sushi place in town. How...</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Restaurant</td>\n",
       "      <td>I was really between 3 and 4 stars for this on...</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Hotel</td>\n",
       "      <td>Comfortable bed, good breakfast, fast internet...</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Attraction                                            Opinion Polarity\n",
       "0  Restaurant  Family diner. Had the buffet. Eclectic assortm...      3.0\n",
       "0  Restaurant  The bun makes the Sonoran Dog. It's like a snu...      4.0\n",
       "0  Restaurant  Yes, this is the only sushi place in town. How...      4.0\n",
       "0  Restaurant  I was really between 3 and 4 stars for this on...      4.0\n",
       "0       Hotel  Comfortable bed, good breakfast, fast internet...      4.0"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(len(yelp_df))\n",
    "yelp_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 145,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(yelp_df[yelp_df['Attraction'] == 'Hotel']['Polarity'], bins=4)\n",
    "plt.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 146,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(yelp_df[yelp_df['Attraction'] == 'Restaurant']['Polarity'], bins=4)\n",
    "plt.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save dataset\n",
    "# yelp_df.to_csv('data/yelp_df_augmented.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add empty column for translation\n",
    "yelp_df['Translation'] = [None for i in range(len(yelp_df))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "12479 12480 12480 12480\n"
     ]
    }
   ],
   "source": [
    "# Split in 4 datasets\n",
    "yelp_df_1 = yelp_df[:int(len(yelp_df)/4)]\n",
    "yelp_df_2 = yelp_df[int(len(yelp_df)/4):int(len(yelp_df)/2)]\n",
    "yelp_df_3 = yelp_df[int(len(yelp_df)/2):int(len(yelp_df)*3/4)]\n",
    "yelp_df_4 = yelp_df[int(len(yelp_df)*3/4):]\n",
    "\n",
    "# Check length\n",
    "print(len(yelp_df_1), len(yelp_df_2), len(yelp_df_3), len(yelp_df_4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save datasets\n",
    "yelp_df_1.to_csv('data/augmentation/yelp_df_1.csv')\n",
    "yelp_df_2.to_csv('data/augmentation/yelp_df_2.csv')\n",
    "yelp_df_3.to_csv('data/augmentation/yelp_df_3.csv')\n",
    "yelp_df_4.to_csv('data/augmentation/yelp_df_4.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "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>Attraction</th>\n",
       "      <th>Opinion</th>\n",
       "      <th>Polarity</th>\n",
       "      <th>Translation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Restaurant</td>\n",
       "      <td>Family diner. Had the buffet. Eclectic assortm...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Restaurant</td>\n",
       "      <td>The bun makes the Sonoran Dog. It's like a snu...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Restaurant</td>\n",
       "      <td>Yes, this is the only sushi place in town. How...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Restaurant</td>\n",
       "      <td>I was really between 3 and 4 stars for this on...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Hotel</td>\n",
       "      <td>Comfortable bed, good breakfast, fast internet...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Attraction                                            Opinion Polarity  \\\n",
       "0  Restaurant  Family diner. Had the buffet. Eclectic assortm...      3.0   \n",
       "0  Restaurant  The bun makes the Sonoran Dog. It's like a snu...      4.0   \n",
       "0  Restaurant  Yes, this is the only sushi place in town. How...      4.0   \n",
       "0  Restaurant  I was really between 3 and 4 stars for this on...      4.0   \n",
       "0       Hotel  Comfortable bed, good breakfast, fast internet...      4.0   \n",
       "\n",
       "  Translation  \n",
       "0        None  \n",
       "0        None  \n",
       "0        None  \n",
       "0        None  \n",
       "0        None  "
      ]
     },
     "execution_count": 171,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yelp_df_1.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Translation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save augmented data\n",
    "# aug_data.to_csv('data/aug_data.csv', index=False)\n",
    "data = pd.read_csv('data/augmentation/yelp_df_2.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 12480\n"
     ]
    }
   ],
   "source": [
    "# Count translated reviews\n",
    "count_translated = 0\n",
    "for review in data['Translation']:\n",
    "    if type(review) == str:\n",
    "        count_translated += 1\n",
    "\n",
    "print(count_translated, len(data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Esto es una prueba'"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from googletrans import Translator # Translation library\n",
    "translator = Translator()\n",
    "\n",
    "# Example\n",
    "translator.translate('This is a test', src='en', dest='es').text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>Unnamed: 0</th>\n",
       "      <th>Attraction</th>\n",
       "      <th>Opinion</th>\n",
       "      <th>Polarity</th>\n",
       "      <th>Translation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>Hotel</td>\n",
       "      <td>Great location. Clean. Old Florida feel only c...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>Hotel</td>\n",
       "      <td>A comfortable stay! Friendly service &amp; clean r...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>Hotel</td>\n",
       "      <td>Hotel is in a great location, the bed was supe...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>Hotel</td>\n",
       "      <td>Like many others, I had a sub-optimal experien...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>Hotel</td>\n",
       "      <td>This is your standard Hampton Inn - the rates ...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0 Attraction                                            Opinion  \\\n",
       "0           0      Hotel  Great location. Clean. Old Florida feel only c...   \n",
       "1           0      Hotel  A comfortable stay! Friendly service & clean r...   \n",
       "2           0      Hotel  Hotel is in a great location, the bed was supe...   \n",
       "3           0      Hotel  Like many others, I had a sub-optimal experien...   \n",
       "4           0      Hotel  This is your standard Hampton Inn - the rates ...   \n",
       "\n",
       "   Polarity  Translation  \n",
       "0       4.0          NaN  \n",
       "1       4.0          NaN  \n",
       "2       3.0          NaN  \n",
       "3       2.0          NaN  \n",
       "4       4.0          NaN  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Great location. Clean. Old Florida feel only complaint is bed had dip in and not very comfy.\n",
      "Error\n"
     ]
    }
   ],
   "source": [
    "from googletrans import Translator # Translation library\n",
    "from IPython.display import clear_output # clear output\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "\n",
    "translator = Translator() # Create translator object\n",
    "\n",
    "# Example\n",
    "translator.translate('This is a test', src='en', dest='es').text\n",
    "\n",
    "# Load data to translate\n",
    "data = pd.read_csv('data/augmentation/yelp_df_3.csv')\n",
    "\n",
    "# SI HAY UN ERROR, COMENTA EL ANTERIOR Y CARGA DONDE SE QUEDO, AQUI:\n",
    "# data = pd.read_csv('data/augmentation/yelp_df_3_translated.csv')\n",
    "\n",
    "# Translate and save\n",
    "\n",
    "count = 0 # Number of reviews translated\n",
    "for id in data.index:\n",
    "    if type(data.at[id, 'Translation']) != str: # If translation not done yet\n",
    "        opinion = str(data.at[id, 'Opinion'])\n",
    "\n",
    "        if len(opinion) > 5000:  # Translate should be less than 5000 characters\n",
    "            opinion = opinion[:5000].copy()\n",
    "\n",
    "        # Try to translate, otherwise save dataframe in current state\n",
    "        try:\n",
    "            data.at[id, 'Translation'] = translator.translate(opinion, src='en', dest='es').text\n",
    "        except:\n",
    "            data.to_csv('data/augmentation/yelp_df_4_translated.csv', index=False)\n",
    "            print('Error')\n",
    "            break\n",
    "\n",
    "        count += 1 # Update count\n",
    "\n",
    "        # Every 100 translations, show progress and save dataframe\n",
    "        if count % 100 == 0:\n",
    "            # Save augmented data\n",
    "            data.to_csv('data/augmentation/yelp_df_4_translated.csv', index=False)\n",
    "            print(f'{count} reviews translated')\n",
    "            clear_output(wait=True)\n",
    "\n",
    "# If translation was successful, save dataframe\n",
    "data.to_csv('data/augmentation/yelp_df_4_translated.csv', index=False)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "TO-DO:\n",
    "1. Separar augmented dataset a hoteles y restaurantes (solamente fijarnos en esas dos categorias).\n",
    "2. Vamos a hacer dos clasificadores:\n",
    "    - Uno para hoteles y otro para restaurantes.\n",
    "3. Datos originales: Separa desde antes 80/20, 20% para validación. Entrenar multilingual pero evaluar solo en el validation en español."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "files = ['data/augmentation/yelp_df_1_translated.csv',\n",
    "        'data/augmentation/yelp_df_2_translated.csv',\n",
    "        'data/augmentation/yelp_df_3_translated.csv',\n",
    "        'data/augmentation/yelp_df_4_translated.csv']\n",
    "\n",
    "df = pd.read_csv(files[0])\n",
    "for i in range(1, 4):\n",
    "    df = pd.concat([df, pd.read_csv(files[i])])\n",
    "\n",
    "# reset index\n",
    "df = df.reset_index(drop=True)\n",
    "\n",
    "# save dataframe\n",
    "df.drop(columns=['Unnamed: 0'], inplace=True)\n",
    "df.to_csv('data/yelp_translated.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('data/yelp_translated.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "49919"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>Attraction</th>\n",
       "      <th>Opinion</th>\n",
       "      <th>Polarity</th>\n",
       "      <th>Translation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Restaurant</td>\n",
       "      <td>Family diner. Had the buffet. Eclectic assortm...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Cena familiar. Tenía el buffet. Surtido ecléct...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Restaurant</td>\n",
       "      <td>The bun makes the Sonoran Dog. It's like a snu...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>El bollo lo hace el Perro Sonorense. Es como u...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Restaurant</td>\n",
       "      <td>Yes, this is the only sushi place in town. How...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>Sí, este es el único lugar de sushi en la ciud...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Attraction                                            Opinion  Polarity  \\\n",
       "0  Restaurant  Family diner. Had the buffet. Eclectic assortm...       3.0   \n",
       "1  Restaurant  The bun makes the Sonoran Dog. It's like a snu...       4.0   \n",
       "2  Restaurant  Yes, this is the only sushi place in town. How...       4.0   \n",
       "\n",
       "                                         Translation  \n",
       "0  Cena familiar. Tenía el buffet. Surtido ecléct...  \n",
       "1  El bollo lo hace el Perro Sonorense. Es como u...  \n",
       "2  Sí, este es el único lugar de sushi en la ciud...  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([5146., 5076., 4687., 3411.]),\n",
       " array([1.  , 1.75, 2.5 , 3.25, 4.  ]),\n",
       " <BarContainer object of 4 artists>)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAD7CAYAAACCEpQdAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAV9ElEQVR4nO3dcUzU9+H/8dfJIetKlw1zV8jFuMV2scGsmJJ27I9jNhncPC+4W7c53Mhm0qjpMHENHQMCwa0rdVRSo5hma1xKzBZ0FZRczybtQrLiVse2Ohe2NE5MBXMcFCdgOfDu8/1jP+8nUj3OeofH+/n4h37en8+H+7x5m6fXj9ydzbIsSwAAoyxb7AsAAKQf8QcAAxF/ADAQ8QcAAxF/ADAQ8QcAAxF/ADCQfbEvYKHGx6cUiyX/koQVK3I1NjaZgitKP+Zyb1oqc1kq85CYiyQtW2bT5z53/y33Z0z8YzHrjuJ//dylgrncm5bKXJbKPCTmkgi3fQDAQMQfAAxE/AHAQMQfAAxE/AHAQMQfAAxE/AHAQBnze/53amY2KofjgcW+jLsmHXOZjlzTxJWPUv44ABbPko//8uws+Z7tXuzLyCgnXqrQxGJfBICUWlD8v//97+vDDz+U3f6/w3fv3q2pqSm98MILikQi+vrXv65du3ZJkgYGBlRfX6+pqSkVFxerublZdrtdw8PDqqmp0djYmL7whS+otbVV999/65ceAwBSJ2H8LcvS4OCg/vCHP8TjPz09LY/Ho46ODhUUFGjbtm3q7e1VaWmpampq9POf/1xFRUWqq6tTZ2enKisr1dzcrMrKSnm9Xh04cEDt7e2qqalJ+QSRvHTdKlsqt+NmZqOLfQlA0hLG/z//+Y8kaevWrbp8+bK+/e1v64tf/KJWrVqllStXSpJ8Pp+CwaAeeughTU9Pq6ioSJLk9/u1b98+fetb39Lp06d14MCB+Pj3vvc94n+P4lZZck68VLHYlwAkLeFv+1y5ckUlJSU6cOCAfvOb3+h3v/udhoeH5XA44sc4nU6FQiGNjIzMGXc4HAqFQhofH1dubm78/xyujwMAFkfCZ/7r1q3TunXr4ttPPfWU9u3bp8ceeyw+ZlmWbDabYrGYbDbbvPHrX29083YiK1bkJnU8kE5L5RbWUpmHxFwSSRj/v/zlL5qdnVVJSYmk/wXd5XIpHA7HjwmHw3I6ncrPz58zPjo6KqfTqby8PE1MTCgajSorKyt+fDLGxibv6G1Nl9IfANy7wuHM//0oh+OBJTEPiblI/3s//9s9aU5422diYkJ79uxRJBLR5OSkjh07ph//+Mc6f/68Lly4oGg0qp6eHrndbrlcLuXk5Ki/v1+S1N3dLbfbrezsbBUXFysQCEiSurq65Ha7k54MAODuSPjMf/369Xrvvfe0adMmxWIxVVZWat26dWppaVF1dbUikYhKS0vl8XgkSa2trWpoaNDk5KQKCwtVVVUlSWpqalJtba0OHjyogoIC7d27N7UzAwDcks2yrIz4uJtPctuH31xJzomXKviZJeHESxVL4hYDt0ruTYt22wcAsPQQfwAwEPEHAAMRfwAwEPEHAAMRfwAwEPEHAAMRfwAw0JL/JC8g1ZbSR4XyMaHmIP7AJ8TnHySHjwm9N3DbBwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEALjv+LL76o2tpaSVJfX598Pp/KysrU1tYWP2ZgYEB+v1/l5eWqr6/XtWvXJEnDw8PasmWLPB6PduzYoampqbs8DQBAMhYU/1OnTunYsWOSpOnpadXV1am9vV2BQEBnz55Vb2+vJKmmpkaNjY06efKkLMtSZ2enJKm5uVmVlZUKBoNau3at2tvbUzQdAMBCJIz/5cuX1dbWpu3bt0uSzpw5o1WrVmnlypWy2+3y+XwKBoMaGhrS9PS0ioqKJEl+v1/BYFCzs7M6ffq0ysvL54wDABaPPdEBjY2N2rVrly5duiRJGhkZkcPhiO93Op0KhULzxh0Oh0KhkMbHx5Wbmyu73T5nPFkrVuQmfQ6Ae5PD8cCSeIx0ScVcbhv/I0eOqKCgQCUlJXr99dclSbFYTDabLX6MZVmy2Wy3HL/+9UY3by/E2NikYjEr6fOW0h8AYKkIhydS+v0djgdS/hjpcqdzWbbMdtsnzbeNfyAQUDgcVkVFhf773//q6tWrGhoaUlZWVvyYcDgsp9Op/Px8hcPh+Pjo6KicTqfy8vI0MTGhaDSqrKys+PEAgMVz23v+hw4dUk9Pj7q7u7Vz5049+eST+vWvf63z58/rwoULikaj6unpkdvtlsvlUk5Ojvr7+yVJ3d3dcrvdys7OVnFxsQKBgCSpq6tLbrc79TMDANxSwnv+N8vJyVFLS4uqq6sViURUWloqj8cjSWptbVVDQ4MmJydVWFioqqoqSVJTU5Nqa2t18OBBFRQUaO/evXd3FgCApCw4/n6/X36/X5JUUlKi48ePzztmzZo1Onr06Lxxl8uljo6OT3CZAIC7iVf4AoCBiD8AGIj4A4CBiD8AGIj4A4CBiD8AGIj4A4CBiD8AGIj4A4CBiD8AGIj4A4CBiD8AGIj4A4CBiD8AGIj4A4CBiD8AGIj4A4CBiD8AGCjpz/AFgE9iZjYqh+OBlD9OOh4jHWZmoyn5vsQfQFotz86S79nuxb6MjHHipYqUfF9u+wCAgYg/ABiI+AOAgYg/ABiI+AOAgYg/ABiI+AOAgYg/ABiI+AOAgYg/ABiI+AOAgRYU/5dfflkbNmyQ1+vVoUOHJEl9fX3y+XwqKytTW1tb/NiBgQH5/X6Vl5ervr5e165dkyQNDw9ry5Yt8ng82rFjh6amplIwHQDAQiSM/7vvvqs//elPOn78uH7/+9+ro6ND//rXv1RXV6f29nYFAgGdPXtWvb29kqSamho1Njbq5MmTsixLnZ2dkqTm5mZVVlYqGAxq7dq1am9vT+3MAAC3lDD+jz/+uF577TXZ7XaNjY0pGo3qypUrWrVqlVauXCm73S6fz6dgMKihoSFNT0+rqKhIkuT3+xUMBjU7O6vTp0+rvLx8zjgAYHEs6LZPdna29u3bJ6/Xq5KSEo2MjMjhcMT3O51OhUKheeMOh0OhUEjj4+PKzc2V3W6fMw4AWBwLfj//nTt36umnn9b27ds1ODgom80W32dZlmw2m2Kx2MeOX/96o5u3E1mxIjep4wFgqUjFB9MkjP+5c+c0MzOjRx55RPfdd5/KysoUDAaVlZUVPyYcDsvpdCo/P1/hcDg+Pjo6KqfTqby8PE1MTCgajSorKyt+fDLGxiYVi1lJnSMtnU/zAWCucHgi6XOWLbPd9klzwts+Fy9eVENDg2ZmZjQzM6O33npLmzdv1vnz53XhwgVFo1H19PTI7XbL5XIpJydH/f39kqTu7m653W5lZ2eruLhYgUBAktTV1SW32530ZAAAd0fCZ/6lpaU6c+aMNm3apKysLJWVlcnr9SovL0/V1dWKRCIqLS2Vx+ORJLW2tqqhoUGTk5MqLCxUVVWVJKmpqUm1tbU6ePCgCgoKtHfv3tTODABwSwu6519dXa3q6uo5YyUlJTp+/Pi8Y9esWaOjR4/OG3e5XOro6LjDywQA3E28whcADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADET8AcBAxB8ADLSg+O/fv19er1der1d79uyRJPX19cnn86msrExtbW3xYwcGBuT3+1VeXq76+npdu3ZNkjQ8PKwtW7bI4/Fox44dmpqaSsF0AAALkTD+fX19+uMf/6hjx46pq6tL//znP9XT06O6ujq1t7crEAjo7Nmz6u3tlSTV1NSosbFRJ0+elGVZ6uzslCQ1NzersrJSwWBQa9euVXt7e2pnBgC4pYTxdzgcqq2t1fLly5Wdna3Vq1drcHBQq1at0sqVK2W32+Xz+RQMBjU0NKTp6WkVFRVJkvx+v4LBoGZnZ3X69GmVl5fPGQcALI6E8X/44YfjMR8cHNQbb7whm80mh8MRP8bpdCoUCmlkZGTOuMPhUCgU0vj4uHJzc2W32+eMAwAWh32hB77//vvatm2bnnvuOWVlZWlwcDC+z7Is2Ww2xWIx2Wy2eePXv97o5u1EVqzITep4AFgqHI4H7vr3XFD8+/v7tXPnTtXV1cnr9erdd99VOByO7w+Hw3I6ncrPz58zPjo6KqfTqby8PE1MTCgajSorKyt+fDLGxiYVi1lJnSOl5ocGAOkUDk8kfc6yZbbbPmlOeNvn0qVLeuaZZ9Ta2iqv1ytJevTRR3X+/HlduHBB0WhUPT09crvdcrlcysnJUX9/vySpu7tbbrdb2dnZKi4uViAQkCR1dXXJ7XYnPRkAwN2R8Jn/q6++qkgkopaWlvjY5s2b1dLSourqakUiEZWWlsrj8UiSWltb1dDQoMnJSRUWFqqqqkqS1NTUpNraWh08eFAFBQXau3dviqYEAEgkYfwbGhrU0NDwsfuOHz8+b2zNmjU6evTovHGXy6WOjo47uEQAwN3GK3wBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEDEHwAMRPwBwEALiv/k5KQ2btyoixcvSpL6+vrk8/lUVlamtra2+HEDAwPy+/0qLy9XfX29rl27JkkaHh7Wli1b5PF4tGPHDk1NTaVgKgCAhUoY//fee0/f/e53NTg4KEmanp5WXV2d2tvbFQgEdPbsWfX29kqSampq1NjYqJMnT8qyLHV2dkqSmpubVVlZqWAwqLVr16q9vT11MwIAJJQw/p2dnWpqapLT6ZQknTlzRqtWrdLKlStlt9vl8/kUDAY1NDSk6elpFRUVSZL8fr+CwaBmZ2d1+vRplZeXzxkHACwee6IDnn/++TnbIyMjcjgc8W2n06lQKDRv3OFwKBQKaXx8XLm5ubLb7XPGAQCLJ2H8bxaLxWSz2eLblmXJZrPdcvz61xvdvL0QK1bkJn0OACwFDscDd/17Jh3//Px8hcPh+HY4HJbT6Zw3Pjo6KqfTqby8PE1MTCgajSorKyt+fLLGxiYVi1lJn5eKHxoApFM4PJH0OcuW2W77pDnpX/V89NFHdf78eV24cEHRaFQ9PT1yu91yuVzKyclRf3+/JKm7u1tut1vZ2dkqLi5WIBCQJHV1dcntdic9EQDA3ZP0M/+cnBy1tLSourpakUhEpaWl8ng8kqTW1lY1NDRocnJShYWFqqqqkiQ1NTWptrZWBw8eVEFBgfbu3Xt3ZwEASMqC4//222/H/7ukpETHjx+fd8yaNWt09OjReeMul0sdHR13eIkAgLuNV/gCgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYiPgDgIGIPwAYKK3xP3HihDZs2KCysjIdPnw4nQ8NALiBPV0PFAqF1NbWptdff13Lly/X5s2b9cQTT+ihhx5K1yUAAP6ftMW/r69PX/7yl/XZz35WklReXq5gMKgf/ehHCzp/2TLbHT+283P33fG5puJnlhx+Xsnh55WcO+lfonNslmVZd3pByXjllVd09epV7dq1S5J05MgRnTlzRj/72c/S8fAAgBuk7Z5/LBaTzfb//yayLGvONgAgfdIW//z8fIXD4fh2OByW0+lM18MDAG6Qtvh/5Stf0alTp/Thhx/qo48+0ptvvim3252uhwcA3CBt/+D74IMPateuXaqqqtLs7KyeeuopfelLX0rXwwMAbpC2f/AFANw7eIUvABiI+AOAgYg/ABiI+AOAgZZU/CcnJ7Vx40ZdvHhx3r6BgQH5/X6Vl5ervr5e165dW4QrXLjbzWX//v1av369KioqVFFRcc++Sd7+/fvl9Xrl9Xq1Z8+eefszaU0SzSVT1kSSXn75ZW3YsEFer1eHDh2atz+T1iXRXDJpXSTpxRdfVG1t7bzxlKyJtUT8/e9/tzZu3GgVFhZaH3zwwbz9Xq/X+tvf/mZZlmX99Kc/tQ4fPpzmK1y4RHPZtm2b9de//nURrmzh3nnnHes73/mOFYlErJmZGauqqsp688035xyTKWuykLlkwppYlmX9+c9/tjZv3mzNzs5aH330kbV+/Xrr3Llzc47JlHVZyFwyZV0sy7L6+vqsJ554wvrJT34yb18q1mTJPPPv7OxUU1PTx75qeGhoSNPT0yoqKpIk+f1+BYPBNF/hwt1uLpJ09uxZvfLKK/L5fNq9e7cikUiarzAxh8Oh2tpaLV++XNnZ2Vq9erWGh4fj+zNpTRLNRcqMNZGkxx9/XK+99prsdrvGxsYUjUb16U9/Or4/k9Yl0VykzFmXy5cvq62tTdu3b5+3L1VrsmTi//zzz6u4uPhj942MjMjhcMS3HQ6HQqFQui4tabeby9TUlB555BHV1NTo2LFjunLlitrb29N8hYk9/PDD8T+sg4ODeuONN1RaWhrfn0lrkmgumbIm12VnZ2vfvn3yer0qKSnRgw8+GN+XSesi3X4umbQujY2N2rVrlz7zmc/M25eqNVky8b+dpfSmcvfff79+9atfafXq1bLb7dq6dat6e3sX+7Ju6f3339fWrVv13HPP6fOf/3x8PBPX5FZzybQ1kaSdO3fq1KlTunTpkjo7O+Pjmbgut5pLpqzLkSNHVFBQoJKSko/dn6o1MSL+N7+p3OjoaMa+qdzw8LCOHj0a37YsS3Z72t6lIyn9/f36wQ9+oGeffVbf+MY35uzLtDW53VwyaU3OnTungYEBSdJ9992nsrIy/fvf/47vz6R1STSXTFmXQCCgd955RxUVFdq3b5/efvtt/eIXv4jvT9WaGBF/l8ulnJwc9ff3S5K6u7sz9k3lPvWpT+mXv/ylPvjgA1mWpcOHD+trX/vaYl/WPJcuXdIzzzyj1tZWeb3eefszaU0SzSVT1kSSLl68qIaGBs3MzGhmZkZvvfWWHnvssfj+TFqXRHPJlHU5dOiQenp61N3drZ07d+rJJ59UXV1dfH+q1mRJx//pp5/WP/7xD0lSa2urXnjhBXk8Hl29elVVVVWLfHXJuT6XvLw87d69Wzt27JDH45FlWfrhD3+42Jc3z6uvvqpIJKKWlpb4r9n99re/zcg1STSXTFkTSSotLdVXv/pVbdq0Sd/85je1bt06eb3ejFyXRHPJpHX5OKleE97YDQAMtKSf+QMAPh7xBwADEX8AMBDxBwADEX8AMBDxBwADEX8AMBDxBwAD/R/Efso6UvBgegAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(df[df['Attraction'] == 'Restaurant']['Polarity'], bins=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([11264.,  5343.,  5634.,  9358.]),\n",
       " array([1.  , 1.75, 2.5 , 3.25, 4.  ]),\n",
       " <BarContainer object of 4 artists>)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(df[df['Attraction'] == 'Hotel']['Polarity'], bins=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_excel('data/train_sentiment_analysis.xlsx')\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Merge Title and Opinion columns\n",
    "for i in range(len(data)):\n",
    "    data.loc[i, 'Opinion'] = str(data.loc[i, 'Title']) + '. ' + str(data.loc[i, 'Opinion'])\n",
    "\n",
    "# Drop the Title column\n",
    "data = data.drop(columns=['Title'])\n",
    "\n",
    "# Drop rows with NAN in Opinion column\n",
    "data = data.dropna(subset=['Opinion'])\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save data in CSV file\n",
    "data.to_csv('data/train_data.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save augmented data\n",
    "# aug_data.to_csv('data/aug_data.csv', index=False)\n",
    "data2 = pd.read_csv('data/augmentation/yelp_df_augmented.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x900 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(4,3))\n",
    "\n",
    "ax.hist(data2['Polarity'], bins=[0.001, 1.001, 2.001, 3.001, 4.001, 5.001], color = '#a2ad59')\n",
    "ax.set_title('Polarity distribution - Yelp data.')\n",
    "ax.set_xlabel('Polarity')\n",
    "ax.set_ylabel('Number of reviews')\n",
    "ax.grid(True)\n",
    "ax.set_xlim([-0.1, 4.1])\n",
    "ax.set_yticks([0, 7000, 14000], [0, '7k', '14k'])\n",
    "ax.set_xticks([0.5, 1.5, 2.5, 3.5], [1, 2, 3, 4])\n",
    "plt.tight_layout()\n",
    "plt.savefig('images/polarity_distribution_yelp.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x900 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(4,3))\n",
    "x = [np.array(data['Polarity'].values), np.array(data2['Polarity'].values)]\n",
    "\n",
    "ax.hist(x, bins=[0.001, 1.001, 2.001, 3.001, 4.001, 5.001],\n",
    "        color = [ '#f98948', '#a2ad59'], stacked=True)\n",
    "\n",
    "ax.set_title('Polarity distribution - Final data.')\n",
    "ax.set_xlabel('Polarity')\n",
    "ax.set_ylabel('Number of reviews')\n",
    "ax.grid(True)\n",
    "ax.set_yticks([0, 10000, 20000], [0, '10k', '20k'])\n",
    "ax.set_xticks([0.5, 1.5, 2.5, 3.5, 4.5], [1, 2, 3, 4, 5])\n",
    "plt.tight_layout()\n",
    "plt.savefig('images/polarity_distribution_total.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "ae0db2a7f29156f5639b4fc85b59b64440ab0d9df8916869e7c653ad5064eeb8"
  },
  "kernelspec": {
   "display_name": "Python 3.10.2 ('nlp')",
   "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": "3.10.2"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}