A04 - Angelo Rosace.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Assignment 4\n",
    "\n",
    "## Angelo Rosace"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 1st Assignment statement\n",
    "\n",
    "Airbnb offers public data about its activities. We will focus on the file listing the available apartments in Barcelona in 2019.\n",
    "\n",
    "We want to make a Naive Bayes Classifier that allow us to tell the neighbour where an apartment could be based on its latitude and longitude coordinates, to assign them to some apartments were the owner forget to upload it.\n",
    "To test if its working just train it with 1000 apartments randomly chosen within the whole file, and test it for another different 1000. Which is the accuracy of your predictions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import random as ra\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.stats import norm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to takle the problem proposed by the assignment we would first have to load our data into a list and inspect them. Inspecting them will allow us to understand how the data i structured and will shed a light on how to handle them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>id</th>\n",
       "      <th>name</th>\n",
       "      <th>host_id</th>\n",
       "      <th>host_name</th>\n",
       "      <th>neighbourhood_group</th>\n",
       "      <th>neighbourhood</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>room_type</th>\n",
       "      <th>price</th>\n",
       "      <th>minimum_nights</th>\n",
       "      <th>number_of_reviews</th>\n",
       "      <th>last_review</th>\n",
       "      <th>reviews_per_month</th>\n",
       "      <th>calculated_host_listings_count</th>\n",
       "      <th>availability_365</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>18666</td>\n",
       "      <td>Flat with Sunny Terrace</td>\n",
       "      <td>71615</td>\n",
       "      <td>Mireia And Maria</td>\n",
       "      <td>Sant Martí</td>\n",
       "      <td>el Camp de l'Arpa del Clot</td>\n",
       "      <td>41.40889</td>\n",
       "      <td>2.18555</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>130</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>2015-10-10</td>\n",
       "      <td>0.02</td>\n",
       "      <td>30</td>\n",
       "      <td>304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>18674</td>\n",
       "      <td>Huge flat for 8 people close to Sagrada Familia</td>\n",
       "      <td>71615</td>\n",
       "      <td>Mireia And Maria</td>\n",
       "      <td>Eixample</td>\n",
       "      <td>la Sagrada Família</td>\n",
       "      <td>41.40420</td>\n",
       "      <td>2.17306</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>60</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>2019-10-19</td>\n",
       "      <td>0.25</td>\n",
       "      <td>30</td>\n",
       "      <td>312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>23197</td>\n",
       "      <td>FORUM DELUXE 5 MINS WALK  CCIB CENTER &amp; SEA!</td>\n",
       "      <td>90417</td>\n",
       "      <td>Etain (Marnie)</td>\n",
       "      <td>Sant Martí</td>\n",
       "      <td>el Besòs i el Maresme</td>\n",
       "      <td>41.41203</td>\n",
       "      <td>2.22114</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>210</td>\n",
       "      <td>3</td>\n",
       "      <td>51</td>\n",
       "      <td>2019-09-29</td>\n",
       "      <td>0.48</td>\n",
       "      <td>2</td>\n",
       "      <td>359</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>25786</td>\n",
       "      <td>NICE ROOM AVAILABLE IN THE HEART OF GRACIA</td>\n",
       "      <td>108310</td>\n",
       "      <td>Pedro</td>\n",
       "      <td>Gràcia</td>\n",
       "      <td>la Vila de Gràcia</td>\n",
       "      <td>41.40145</td>\n",
       "      <td>2.15645</td>\n",
       "      <td>Private room</td>\n",
       "      <td>32</td>\n",
       "      <td>1</td>\n",
       "      <td>268</td>\n",
       "      <td>2019-11-06</td>\n",
       "      <td>2.38</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>31958</td>\n",
       "      <td>Gaudi Attic 1, Friendly Rentals</td>\n",
       "      <td>136853</td>\n",
       "      <td>Fidelio</td>\n",
       "      <td>Gràcia</td>\n",
       "      <td>el Camp d'en Grassot i Gràcia Nova</td>\n",
       "      <td>41.40950</td>\n",
       "      <td>2.15938</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>60</td>\n",
       "      <td>1</td>\n",
       "      <td>182</td>\n",
       "      <td>2019-10-16</td>\n",
       "      <td>1.71</td>\n",
       "      <td>39</td>\n",
       "      <td>324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20423</td>\n",
       "      <td>39968822</td>\n",
       "      <td>Bright, sunshine white sophisticated</td>\n",
       "      <td>285303484</td>\n",
       "      <td>Ann</td>\n",
       "      <td>Horta-Guinardó</td>\n",
       "      <td>el Baix Guinardó</td>\n",
       "      <td>41.41177</td>\n",
       "      <td>2.16373</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>50</td>\n",
       "      <td>32</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>349</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20424</td>\n",
       "      <td>39969034</td>\n",
       "      <td>Fabulous, Modern Barcelona 2 Bedroom Apartment...</td>\n",
       "      <td>268096872</td>\n",
       "      <td>Allianz</td>\n",
       "      <td>Ciutat Vella</td>\n",
       "      <td>el Barri Gòtic</td>\n",
       "      <td>41.38576</td>\n",
       "      <td>2.17438</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>508</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11</td>\n",
       "      <td>362</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20425</td>\n",
       "      <td>39969285</td>\n",
       "      <td>Spectacular Luxurious 6 Bedroom Penthouse w/ P...</td>\n",
       "      <td>268096872</td>\n",
       "      <td>Allianz</td>\n",
       "      <td>Eixample</td>\n",
       "      <td>la Sagrada Família</td>\n",
       "      <td>41.40758</td>\n",
       "      <td>2.17242</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>1031</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11</td>\n",
       "      <td>362</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20426</td>\n",
       "      <td>39972885</td>\n",
       "      <td>Couples` Home with Parking</td>\n",
       "      <td>252088393</td>\n",
       "      <td>Joseph R.</td>\n",
       "      <td>Ciutat Vella</td>\n",
       "      <td>Sant Pere, Santa Caterina i la Ribera</td>\n",
       "      <td>41.39138</td>\n",
       "      <td>2.18240</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>89</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20427</td>\n",
       "      <td>39973807</td>\n",
       "      <td>Un lugar de descanso para tus vacaciones</td>\n",
       "      <td>303423208</td>\n",
       "      <td>Juanito</td>\n",
       "      <td>Eixample</td>\n",
       "      <td>la Sagrada Família</td>\n",
       "      <td>41.40357</td>\n",
       "      <td>2.17408</td>\n",
       "      <td>Private room</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>362</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20428 rows × 16 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             id                                               name    host_id  \\\n",
       "0         18666                            Flat with Sunny Terrace      71615   \n",
       "1         18674    Huge flat for 8 people close to Sagrada Familia      71615   \n",
       "2         23197       FORUM DELUXE 5 MINS WALK  CCIB CENTER & SEA!      90417   \n",
       "3         25786         NICE ROOM AVAILABLE IN THE HEART OF GRACIA     108310   \n",
       "4         31958                    Gaudi Attic 1, Friendly Rentals     136853   \n",
       "...         ...                                                ...        ...   \n",
       "20423  39968822               Bright, sunshine white sophisticated  285303484   \n",
       "20424  39969034  Fabulous, Modern Barcelona 2 Bedroom Apartment...  268096872   \n",
       "20425  39969285  Spectacular Luxurious 6 Bedroom Penthouse w/ P...  268096872   \n",
       "20426  39972885                         Couples` Home with Parking  252088393   \n",
       "20427  39973807           Un lugar de descanso para tus vacaciones  303423208   \n",
       "\n",
       "              host_name neighbourhood_group  \\\n",
       "0      Mireia And Maria          Sant Martí   \n",
       "1      Mireia And Maria            Eixample   \n",
       "2        Etain (Marnie)          Sant Martí   \n",
       "3                 Pedro              Gràcia   \n",
       "4               Fidelio              Gràcia   \n",
       "...                 ...                 ...   \n",
       "20423               Ann      Horta-Guinardó   \n",
       "20424           Allianz        Ciutat Vella   \n",
       "20425           Allianz            Eixample   \n",
       "20426         Joseph R.        Ciutat Vella   \n",
       "20427           Juanito            Eixample   \n",
       "\n",
       "                               neighbourhood  latitude  longitude  \\\n",
       "0                 el Camp de l'Arpa del Clot  41.40889    2.18555   \n",
       "1                         la Sagrada Família  41.40420    2.17306   \n",
       "2                      el Besòs i el Maresme  41.41203    2.22114   \n",
       "3                          la Vila de Gràcia  41.40145    2.15645   \n",
       "4         el Camp d'en Grassot i Gràcia Nova  41.40950    2.15938   \n",
       "...                                      ...       ...        ...   \n",
       "20423                       el Baix Guinardó  41.41177    2.16373   \n",
       "20424                         el Barri Gòtic  41.38576    2.17438   \n",
       "20425                     la Sagrada Família  41.40758    2.17242   \n",
       "20426  Sant Pere, Santa Caterina i la Ribera  41.39138    2.18240   \n",
       "20427                     la Sagrada Família  41.40357    2.17408   \n",
       "\n",
       "             room_type  price  minimum_nights  number_of_reviews last_review  \\\n",
       "0      Entire home/apt    130               3                  1  2015-10-10   \n",
       "1      Entire home/apt     60               1                 20  2019-10-19   \n",
       "2      Entire home/apt    210               3                 51  2019-09-29   \n",
       "3         Private room     32               1                268  2019-11-06   \n",
       "4      Entire home/apt     60               1                182  2019-10-16   \n",
       "...                ...    ...             ...                ...         ...   \n",
       "20423  Entire home/apt     50              32                  0         NaN   \n",
       "20424  Entire home/apt    508               3                  0         NaN   \n",
       "20425  Entire home/apt   1031               3                  0         NaN   \n",
       "20426  Entire home/apt     89               3                  0         NaN   \n",
       "20427     Private room     50               1                  0         NaN   \n",
       "\n",
       "       reviews_per_month  calculated_host_listings_count  availability_365  \n",
       "0                   0.02                              30               304  \n",
       "1                   0.25                              30               312  \n",
       "2                   0.48                               2               359  \n",
       "3                   2.38                               1                46  \n",
       "4                   1.71                              39               324  \n",
       "...                  ...                             ...               ...  \n",
       "20423                NaN                               1               349  \n",
       "20424                NaN                              11               362  \n",
       "20425                NaN                              11               362  \n",
       "20426                NaN                               1                22  \n",
       "20427                NaN                               3               362  \n",
       "\n",
       "[20428 rows x 16 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "appartments = pd.read_csv(\"listings.csv\", header = [0], sep = \",\")\n",
    "appartments"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The assignment asks us to just take in consideration altitude and longitude and neighborhood, therefore I will create a subset of the data that just holds informatin for this three variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>neighbourhood_group</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Sant Martí</td>\n",
       "      <td>41.40889</td>\n",
       "      <td>2.18555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Eixample</td>\n",
       "      <td>41.40420</td>\n",
       "      <td>2.17306</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Sant Martí</td>\n",
       "      <td>41.41203</td>\n",
       "      <td>2.22114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Gràcia</td>\n",
       "      <td>41.40145</td>\n",
       "      <td>2.15645</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Gràcia</td>\n",
       "      <td>41.40950</td>\n",
       "      <td>2.15938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20423</td>\n",
       "      <td>Horta-Guinardó</td>\n",
       "      <td>41.41177</td>\n",
       "      <td>2.16373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20424</td>\n",
       "      <td>Ciutat Vella</td>\n",
       "      <td>41.38576</td>\n",
       "      <td>2.17438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20425</td>\n",
       "      <td>Eixample</td>\n",
       "      <td>41.40758</td>\n",
       "      <td>2.17242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20426</td>\n",
       "      <td>Ciutat Vella</td>\n",
       "      <td>41.39138</td>\n",
       "      <td>2.18240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20427</td>\n",
       "      <td>Eixample</td>\n",
       "      <td>41.40357</td>\n",
       "      <td>2.17408</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20428 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      neighbourhood_group  latitude  longitude\n",
       "0              Sant Martí  41.40889    2.18555\n",
       "1                Eixample  41.40420    2.17306\n",
       "2              Sant Martí  41.41203    2.22114\n",
       "3                  Gràcia  41.40145    2.15645\n",
       "4                  Gràcia  41.40950    2.15938\n",
       "...                   ...       ...        ...\n",
       "20423      Horta-Guinardó  41.41177    2.16373\n",
       "20424        Ciutat Vella  41.38576    2.17438\n",
       "20425            Eixample  41.40758    2.17242\n",
       "20426        Ciutat Vella  41.39138    2.18240\n",
       "20427            Eixample  41.40357    2.17408\n",
       "\n",
       "[20428 rows x 3 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "appartment_useful_prop = appartments[[\"neighbourhood_group\",\"latitude\",\"longitude\"]].copy()\n",
    "appartment_useful_prop"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once we created our data set of interest we can go over to implement the arlogrithm. First of all I will implement a class describing the object *Appartment*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Appartment( object ):\n",
    "    \n",
    "    def __init__(self, latitude, longitude, neighbourhood):\n",
    "        self.latitude=latitude\n",
    "        self.longitude=longitude\n",
    "        self.neighbourhood=neighbourhood\n",
    "    \n",
    "    def getFeaturesDict(self): return {\"latitude\":self.latitude, \"longitude\": self.longitude, \"neighbourhood\": self.neighbourhood}\n",
    "    \n",
    "    def getNeighbourhood(self): return self.neighbourhood\n",
    "    \n",
    "    def getLatitude(self): return self.latitude\n",
    "    \n",
    "    def getLongitude(self): return self.longitude"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that I create an object implementation of the class Appartment I can convert my data into objects and store them in a list."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "to_objs = [list(appartment_useful_prop.loc[i]) for i in range(len(appartment_useful_prop))]\n",
    "objs = [Appartment(item[1],item[2],item[0]) for item in to_objs]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To comply with the statement of the assignment I will use a training set of 1000 appartments and test the model against other 1000 appartments. I will therefore generate both training and testing set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_set = ra.sample(objs,k=1000)\n",
    "test_set = ra.sample(objs,k=1000)\n",
    "test_set_without_neighbourhood = [[i.getLatitude(), i.getLongitude()] for i in test_set]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The way the elements to put in the different sets where chosen was supposely random. For both sets a random sample of 1000 elements has been selected.  \n",
    "\n",
    "Now that our two sets are ready we can finally go over to implementing the naive bayes algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "class NaiveBayesOfAppartments( object ):\n",
    " \n",
    "    def _distributionProbFunction(self,neighbourhood, Property):\n",
    "        # Return the result of evaluating a Normal distribution of:\n",
    "        #   mean equal to the mean of the property for this type of appartment\n",
    "        #   std  equal to the std  of the property for this type of appartment\n",
    "        return lambda x: norm.pdf( x, self.cmean[neighbourhood][Property],self.cstdv[neighbourhood][Property] )\n",
    "        \n",
    "    def __init__(self, appartments):\n",
    "        self.appartments = appartments\n",
    " \n",
    "        self.neighbourhoods = {}\n",
    "        for appartment in self.appartments:\n",
    "            if appartment.getNeighbourhood() in self.neighbourhoods.keys():\n",
    "                self.neighbourhoods[appartment.getNeighbourhood()] += 1\n",
    "            else: self.neighbourhoods[appartment.getNeighbourhood()] = 1\n",
    " \n",
    "        self.probl = {}\n",
    "        self.N = len(self.appartments[0].getFeaturesDict())\n",
    "        self.properties = list(self.appartments[0].getFeaturesDict().keys())\n",
    "        self.cmean = {}\n",
    "        self.cstdv = {}\n",
    "        self.prob = {}\n",
    "        \n",
    "        for neighbourhood in self.neighbourhoods.keys():\n",
    "            # The amount of appartments of each type that I have in my \"training\" dataset\n",
    "            self.probl[neighbourhood] = self.neighbourhoods[neighbourhood]/len(self.appartments)\n",
    "            \n",
    "            # I will store the mean, std, and normal probabiblity function,\n",
    "            # of each property for each type of appartments\n",
    "            self.cmean[neighbourhood] = {} #[0 for i in range(self.N)]\n",
    "            self.cstdv[neighbourhood] = {} #[0 for i in range(self.N)]\n",
    "            self.prob[neighbourhood] = {} #[0 for i in range(self.N)]\n",
    "            \n",
    "            # Then, for each property\n",
    "            for i, prop in enumerate(self.properties):\n",
    "                # Make a numpy array taking the list of the read properties\n",
    "                try:\n",
    "                    xi = np.array([float(e.getFeaturesDict()[prop]) for e in self.appartments if e.getNeighbourhood() == neighbourhood])\n",
    "                except:\n",
    "                    #This pass statement has been inserted becaused python would not allow turning the type of the variable to float.\n",
    "                    #Although the e.getFeaturesDict()[prop] value should just be an integer variable, python recognize it as a tuple (int, string). THe string part of the tuple\n",
    "                    #raises the exception. We know that the exception does not make sense because by printing the array xi we actually see that it just contains integer values.\n",
    "                    #that is why we just pass the exception.\n",
    "                    pass\n",
    "                # Fullfill the mean with the mean, the std with the std and the prob with the probability function\n",
    "                self.cmean[neighbourhood][prop] = xi.mean()\n",
    "                self.cstdv[neighbourhood][prop] = xi.std()\n",
    "                self.prob[neighbourhood][prop] = self._distributionProbFunction(neighbourhood,prop)\n",
    "\n",
    "    def predict(self, X):\n",
    "        p = {}\n",
    "        \n",
    "        # In the moment of predicting I have the probabilities of each feature of belonging to each class.\n",
    "        for neighbourhood in self.neighbourhoods.keys():\n",
    "            vectorToMultiply = [self._distributionProbFunction(neighbourhood,prop)(X[i]) for i, prop in enumerate(self.properties[:-1]) if self._distributionProbFunction(neighbourhood,prop)(X[i]) != 0.0 ]\n",
    "            \n",
    "            cp = np.prod( vectorToMultiply )\n",
    "            \n",
    "            p[neighbourhood] = self.probl[neighbourhood]*cp\n",
    "            \n",
    "        results = [(p[neighbourhood],neighbourhood) for neighbourhood in self.neighbourhoods.keys()]\n",
    "        results.sort()\n",
    "        return [results[-1][1], results[-1][0]/sum([r[0] for r in results])]\n",
    "\n",
    "    def plotDistributions(self):\n",
    "        for i, prop in enumerate(self.properties):\n",
    "            nprops=len(range(self.N))\n",
    "            rows = int(nprops/2)+int(nprops%2)\n",
    "            plt.subplot(rows,2,i+1)\n",
    "            for neighbourhood in list(self.neighbourhoods.keys()):\n",
    "                mean = self.cmean[neighbourhood][prop]\n",
    "                var = self.cstdv[neighbourhood][prop]\n",
    "                x= np.linspace(mean-4*var,mean+4*var,200)\n",
    "                xmin,xmax=plt.xlim()\n",
    "                plt.xlim(min(xmin,mean-3*var),max(xmax,mean+3*var))\n",
    "                plt.title(self.properties[i])\n",
    "                \n",
    "                self.prob[neighbourhood][prop] = self._distributionProbFunction(neighbourhood,prop)\n",
    "\n",
    "                plt.plot(x,self.prob[neighbourhood][prop](x), label = neighbourhood)\n",
    "                leg = plt.legend(loc=\"upper left\")\n",
    "                leg.get_frame().set_alpha(0.7)\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once completed the implementation of the algorithm we have to generate our model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Eixample': 0.303, 'Ciutat Vella': 0.256, 'Horta-Guinardó': 0.026, 'Sants-Montjuïc': 0.122, 'Sant Martí': 0.105, 'Sant Andreu': 0.021, 'Gràcia': 0.084, 'Les Corts': 0.026, 'Nou Barris': 0.009, 'Sarrià-Sant Gervasi': 0.048}\n"
     ]
    }
   ],
   "source": [
    "NBC_model = NaiveBayesOfAppartments(train_set)\n",
    "print(NBC_model.probl)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The dictionary outputed above describes us the model that explains our training set and furthermore tells us which is the probability of a flat to be in one or the other neighbourhood group.\n",
    "\n",
    "Now we can use this model to test our testing set.\n",
    "\n",
    "Testing our model will mean using our model to predict the neighbourhood of each appartment in the set. Then we will compare the result with our original set to see if the prediction was correct and to which extent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction = []\n",
    "for appartment in test_set_without_neighbourhood:\n",
    "    prediction.append(NBC_model.predict(appartment))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we have our prediction we will check it against the test_set (with the information about the neighbour group) in order to understand how precise our prediction was."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "By comparing my prediction to the original test_set we can see that my model obtains an accuracy of 89.0 %\n"
     ]
    }
   ],
   "source": [
    "pred_right = 0\n",
    "pred_wrong = 0\n",
    "\n",
    "for i,pred in enumerate(prediction):\n",
    "    if pred[0] == test_set[i].getNeighbourhood():\n",
    "        pred_right += 1\n",
    "    else:\n",
    "        pred_wrong += 1\n",
    "\n",
    "accuracy = (pred_right/(pred_right+pred_wrong))*100\n",
    "print(\"By comparing my prediction to the original test_set we can see that my model obtains an accuracy of\",accuracy,\"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2nd Assignment statement"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We want to clusterize the apartments according to their prize and room type (1: shared room, 2: private room, 3: hotel room, 4: entire apartment). Plotting different amount of clusters according to this and other features (neighbourhood, availability, minimum nights).\n",
    "\n",
    "Can we draw any conclusions?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We already imported the data we need but now the assignment to tackle is about clusterization rather then classification.\n",
    "For this reason we will implement a k-means algorithm.\n",
    "\n",
    "K-means is an algorithm that allow us to clusterize our data according to the variables of the data set we are analysing. For this purpose the approach requires the definition of distances between the elements of our data set. This distances will be calculated by means of the eucledian theorem.\n",
    "\n",
    "We therefore have to implement a function that allow us to compute this distances."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "def vector_euclidian_distance(feat_one, feat_two):\n",
    "    squared_distance = 0\n",
    "    #Assuming correct input to the function where the lengths of two features are the same\n",
    "    for i in range(len(feat_two)-1):\n",
    "            squared_distance += (feat_one[i] - feat_two[i])**2\n",
    "    ed = math.sqrt(squared_distance)\n",
    "    return ed"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we defined the type of distance we will be using in the algorithm let's go over implementing a class for it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "class K_Means:\n",
    "\n",
    "    def __init__(self, k =3, tolerance = 0.0001, max_iterations = 300):\n",
    "        self.k = k\n",
    "        self.tolerance = tolerance\n",
    "        self.max_iterations = max_iterations\n",
    "    \n",
    "\n",
    "    def fit(self, data):\n",
    "        #initialize the centroids, the first 'k' elements in the dataset will be our initial centroids\n",
    "        self.centroids = [None for _ in range(self.k)]\n",
    "        #self.centroids = [i[0:2] for i in data[0:self.k]]\n",
    "        for i in range(self.k):\n",
    "            #self.centroids[i] = data[i]\n",
    "            self.centroids[i] = data[i]\n",
    "\n",
    "        for i in range(self.max_iterations):\n",
    "            self.classes = {c:[] for c in range(self.k)}\n",
    "            for i in range(self.k):\n",
    "                self.classes[i] = []\n",
    "\n",
    "                #find the distance between the point and cluster; choose the nearest centroid\n",
    "                for features in data:\n",
    "                    distances = [vector_euclidian_distance(features, self.centroids[centroid]) for centroid in range(len(self.centroids))]\n",
    "                    classification = distances.index(min(distances))\n",
    "                    self.classes[classification].append(features)\n",
    "            \n",
    "            #previous = dict(self.centroids)\n",
    "            \n",
    "            #average the cluster datapoints to re-calculate the centroids\n",
    "            for classification in self.classes:\n",
    "                self.centroids[classification] = np.average(self.classes[classification], axis = 0)\n",
    "            "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we implemented the k means class we have to prepare the data we want to feed to the algorithm. Once we did that we can cluster our data by means of the k-means algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "clustering_set = appartments.sample(frac=0.1, replace=True)\n",
    "X = clustering_set[\"longitude\"]\n",
    "Y = clustering_set[\"latitude\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once the algorithm has finished processing the data we can finally plot them. We will do it differentiating the elements in turn by price, room type, neighbourhood, availability and minimum nights.\n",
    "\n",
    "**Clusterization by Prize**\n",
    "\n",
    "Let's first of all define a scale for the prices. We know that the maximum and the minimum prices are respectively:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "15000 and 0\n"
     ]
    }
   ],
   "source": [
    "price = appartments[\"price\"]\n",
    "maxP = max(price)\n",
    "print(maxP, \"and\", min(price))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This will be the extremes of our color scale. Let us then define different price levels that will in turn correspond to a different color. We will define 5 levels that represent spans from very low to very high for which we will clusterize the appartments."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "K-mean algorithm ran correctly\n"
     ]
    }
   ],
   "source": [
    "P = clustering_set[\"price\"]\n",
    "new_P = []\n",
    "\n",
    "for i in P:\n",
    "    if i <= 50:\n",
    "        new_P.append(1)\n",
    "    if i > 50 and i <= 300:\n",
    "        new_P.append(2)\n",
    "    if i > 300 and i <= 1000:\n",
    "        new_P.append(3)\n",
    "    if i > 1000 and i <= 2500:\n",
    "        new_P.append(4)\n",
    "    if i > 2500 and i < maxP:\n",
    "        new_P.append(5)\n",
    "\n",
    "Z = list(zip(X,Y,new_P))\n",
    "km = K_Means(5)\n",
    "km.fit(Z)\n",
    "print(\"K-mean algorithm ran correctly\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Centroid: [ 2.17895686 41.38669258]\n",
      "Centroid: [ 2.16171424 41.38127715]\n",
      "Centroid: [ 2.19642783 41.41273327]\n",
      "Centroid: [ 2.13740114 41.3806372 ]\n",
      "Centroid: [ 2.16280476 41.40656177]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors = {\n",
    "    1:\"b\",\n",
    "    2:\"g\",\n",
    "    3:\"y\",\n",
    "    4:\"r\",\n",
    "    5:\"c\",\n",
    "    6:\"k\",\n",
    "    7:\"m\",\n",
    "    8:[106/255, 90/255, 205/255],\n",
    "    9:[151/255, 255/255, 0],\n",
    "    10:[151/255, 122/255, 0]\n",
    "} \n",
    "\n",
    "for centroid in km.centroids:\n",
    "    print (\"Centroid:\", centroid[0:2])\n",
    "\n",
    "for c,classification in enumerate(km.classes):\n",
    "    for p,features in enumerate(km.classes[classification]):\n",
    "        plt.scatter(features[0], features[1], color = colors[features[2]],s = 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Clusterization by room type**\n",
    "\n",
    "We know that we have four room types (1: shared room, 2: private room, 3: hotel room, 4: entire apartment) and in order to use them to clusterize our data we will first have to map those types to numeric values. After having mapped the room types we can go on to run the K-means algorithm and plot the results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "K-mean algorithm ran correctly\n"
     ]
    }
   ],
   "source": [
    "rooms = clustering_set[\"room_type\"]\n",
    "mapped_rooms = []\n",
    "\n",
    "for room in rooms:\n",
    "    if room == \"Shared room\":\n",
    "        mapped_rooms.append(1)\n",
    "    if room == \"Private room\":\n",
    "        mapped_rooms.append(2)\n",
    "    if room == \"Hotel room\":\n",
    "        mapped_rooms.append(3)\n",
    "    if room == \"Entire home/apt\":\n",
    "        mapped_rooms.append(4)\n",
    "        \n",
    "Z = list(zip(X,Y,mapped_rooms))\n",
    "km = K_Means(4)\n",
    "km.fit(Z)\n",
    "print(\"K-mean algorithm ran correctly\") "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Centroid: [ 2.16300001 41.40571411]\n",
      "Centroid: [ 2.17188172 41.38284053]\n",
      "Centroid: [ 2.19427304 41.40951676]\n",
      "Centroid: [ 2.14396154 41.38139271]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for centroid in km.centroids:\n",
    "    print (\"Centroid:\", centroid[0:2])\n",
    "\n",
    "for c,classification in enumerate(km.classes):\n",
    "    for p,features in enumerate(km.classes[classification]):\n",
    "        plt.scatter(features[0], features[1], color = colors[features[2]],s = 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By comparing the plot for the price and this last one we can already see that the central part of Barcelona is where we have the most biggest appartments. By comparing the two previous plots we actually see that size does not directly imply that the appartment is going to be expansive. This can be for the most cases be considered a general rule but we have evidence of some appartments that are expensive although being just a private room. That could be do to the neighbourhood in which that particular room. By further analysing the different varaibles of the data set we will probabliy uncover more underlyng patterns that could maybe explain this initial guess."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Clusterization by neighbourhood group**\n",
    "\n",
    "We know that our data set contains appartments located in eleven different neighbourhood groups. Those are:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Eixample', 'Sants-Montjuïc', 'Sarrià-Sant Gervasi', 'Gràcia', 'Sant Martí', 'Horta-Guinardó', 'Nou Barris', 'Les Corts', 'Ciutat Vella', 'Sant Andreu'}\n"
     ]
    }
   ],
   "source": [
    "neighbourhoods = set(appartments[\"neighbourhood_group\"])\n",
    "print(neighbourhoods)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now want to clusterize our data in the different neighbourhoods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "K-mean algorithm ran correctly\n"
     ]
    }
   ],
   "source": [
    "neighs = clustering_set[\"neighbourhood_group\"]\n",
    "N = []\n",
    "\n",
    "for neigh in neighs:\n",
    "    if neigh == \"Eixample\":\n",
    "        N.append(1)\n",
    "    if neigh == \"Sants-Montjuïc\":\n",
    "        N.append(2)\n",
    "    if neigh == \"Sarrià-Sant Gervasi\":\n",
    "        N.append(3)\n",
    "    if neigh == \"Gràcia\":\n",
    "        N.append(4)\n",
    "    if neigh == \"Sant Martí\":\n",
    "        N.append(5)\n",
    "    if neigh == \"Horta-Guinardó\":\n",
    "        N.append(6)\n",
    "    if neigh == \"Nou Barris\":\n",
    "        N.append(7)\n",
    "    if neigh == \"Les Corts\":\n",
    "        N.append(8)\n",
    "    if neigh == \"Ciutat Vella\":\n",
    "        N.append(9)\n",
    "    if neigh == \"Sant Andreu\":\n",
    "        N.append(10)\n",
    "\n",
    "Z = list(zip(X,Y,N))\n",
    "km = K_Means(4)\n",
    "km.fit(Z)\n",
    "print(\"K-mean algorithm ran correctly\") "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Centroid: [ 2.16300001 41.40571411]\n",
      "Centroid: [ 2.17188172 41.38284053]\n",
      "Centroid: [ 2.19427304 41.40951676]\n",
      "Centroid: [ 2.14396154 41.38139271]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for centroid in km.centroids:\n",
    "    print (\"Centroid:\", centroid[0:2])\n",
    "\n",
    "for c,classification in enumerate(km.classes):\n",
    "    for p,features in enumerate(km.classes[classification]):\n",
    "        plt.scatter(features[0], features[1], color = colors[features[2]],s = 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By comparing the three plots we can se that the most expensive appartments are towards the outskirts of Barcelona which means around the neighbourhoods of Sant Martì, Gracia and Nou Barris."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Clusterization by availability**\n",
    "\n",
    "At the moment in which we downloaded the data some of the appartments where avalible and some not. We want to clusterize the data accordingly to this variable. The varaible tells us how many days a certain appartment is available for this year. In order to clusterize the data we will consider four different ranges of ammounts of avvailable days (up to 1 month, from 1 to 3 moths, from 3 to 6 months, more then 6 moths) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "K-mean algorithm ran correctly\n"
     ]
    }
   ],
   "source": [
    "availability = clustering_set[\"availability_365\"]\n",
    "A = []\n",
    "\n",
    "for av in availability:\n",
    "    if av <= 31:\n",
    "        A.append(1)\n",
    "    if av > 31 and av <= 66:\n",
    "        A.append(2)\n",
    "    if av > 66 and av <= 186:\n",
    "        A.append(3)\n",
    "    if av > 186:\n",
    "        A.append(4)\n",
    "\n",
    "Z = list(zip(X,Y,A))\n",
    "km = K_Means(4)\n",
    "km.fit(Z)\n",
    "print(\"K-mean algorithm ran correctly\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Centroid: [ 2.16300001 41.40571411]\n",
      "Centroid: [ 2.17188172 41.38284053]\n",
      "Centroid: [ 2.19427304 41.40951676]\n",
      "Centroid: [ 2.14396154 41.38139271]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for centroid in km.centroids:\n",
    "    print (\"Centroid:\", centroid[0:2])\n",
    "\n",
    "for c,classification in enumerate(km.classes):\n",
    "    for p,features in enumerate(km.classes[classification]):\n",
    "        plt.scatter(features[0], features[1], color = colors[features[2]],s = 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By looking at this last graph we see a higher concentration of red dots. This means that generally people would try to rent houses for a shorter amount of time, therefore leaving a higher amount of avvailable days. That makes sense if we think that AirBnb is a platform for short-term hause renting."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Clusterization by minimum nights**\n",
    "\n",
    "At the moment in which we downloaded the data we also have some information about the duration of a minimum stay in all the appartments. We want now to cluster our data basing our selves on this variable.\n",
    "By inspecting the data set we see that this particular variable is continuous. Therefore, in order to propperly clusterize the data I will firs inspecct the variable. In particular I want to know which are the minimum and maximum values for the variable. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the variable has values that range from 1 to 365\n"
     ]
    }
   ],
   "source": [
    "nights = clustering_set[\"minimum_nights\"]\n",
    "print(\"the variable has values that range from\", min(nights), \"to\", max(nights))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Taking this in consideration we will define a 4 ranks for the varaibles, exactly as we did for the availability varaible."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "K-mean algorithm ran correctly\n"
     ]
    }
   ],
   "source": [
    "nig = []\n",
    "\n",
    "for n in nights:\n",
    "    if n <= 31:\n",
    "        nig.append(1)\n",
    "    if n > 31 and av <= 66:\n",
    "        nig.append(2)\n",
    "    if n > 66 and av <= 186:\n",
    "        nig.append(3)\n",
    "    if n > 186:\n",
    "        nig.append(4)\n",
    "\n",
    "Z = list(zip(X,Y,nig))\n",
    "km = K_Means(4)\n",
    "km.fit(Z)\n",
    "print(\"K-mean algorithm ran correctly\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Centroid: [ 2.16468541 41.40650835]\n",
      "Centroid: [ 2.17200535 41.38295836]\n",
      "Centroid: [ 2.19691461 41.40958029]\n",
      "Centroid: [ 2.14445391 41.38193834]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for centroid in km.centroids:\n",
    "    print (\"Centroid:\", centroid[0:2])\n",
    "\n",
    "for c,classification in enumerate(km.classes):\n",
    "    for p,features in enumerate(km.classes[classification]):\n",
    "        plt.scatter(features[0], features[1], color = colors[features[2]],s = 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Last but not least we consider this last graph. By looking at it is immediately clear that the most of the appartments or almost all of them are being rented for not more than a month which again makes sense considering the purpose of the service that AirBnb provides."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, by collecting all this information we can conclude that not surprisingly a pricier house is harder to rent for a longer amount of time and that it will likely be placed in the outskirts of Barcelona.\n",
    "Moreover we can spot a tendency to rent full appartments/houses.\n",
    "In general, very few houses are unavailable for more then one month."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}