recursos_proyecto_1 Julio Becerra.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Proyecto: Análisis mercado inmobiliario de la Ciudad de Buenos Aires"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "¡Bienvenidos al primer proyecto de la carrera de Data Science de Acamica! \n",
    "\n",
    "En este proyecto vamos a trabajar con un dataset de propiedades en venta publicadas en el portal [Properati](www.properati.com.ar). El objetivo es dar los primeros pasos en el proceso de exploración de datos usando las herramientas que trabajamos en las activades.\n",
    "\n",
    "Vamos a enfocarnos en analizar algunos aspectos del mercado inmobiliario de la Ciudad de Buenos Aires pero para eso primero debemos limpiar un poco el dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## El dataset: propiedades en venta en Ciudad de Buenos Aires."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "En este proyecto, trabajaremos con una muestra del conjunto de datos de propiedades en venta mencionado. En este dataset, cada fila es una propiedad en venta. \n",
    "\n",
    "## Los atributos\n",
    "A continuación vamos a describir los atributos que consideramos en esta muestra:\n",
    "\n",
    "* **id**: id de la propiedad\n",
    "* **created_on**: fecha en la que la propiedad ingresó al sitio\n",
    "* **operation**: alquiler (rent) o venta (sell)\n",
    "* **property_type**: tipo de propiedad (casa, departamento, ph, etcétera)\n",
    "* **place_with_parent_names**: nombre del lugar donde se encuentra la propiedad según el publicador\n",
    "* **lat-lon**: coordenadas concatenadas\n",
    "* **lat**: latitud\n",
    "* **lon**: longitud\n",
    "* **price**: precio en la moneda especificada en currency\n",
    "* **currency**: divisa en la que está expresada la publicación\n",
    "* **price_aprox_usd**: precio aproximado en dólares estadounidenses\n",
    "* **surface_total_in_m2**: superficie total (en metros cuadrados)\n",
    "* **surface_covered_in_m2**: superficie cubierta (en metros cuadrados)\n",
    "* **price_usd_per_m2**: precio por metro cuadrado en dólares (precio dólares / superficie)\n",
    "* **floor**: número de piso (si corresponde)\n",
    "* **rooms**: cantidad de ambientes\n",
    "* **expenses**: expensas (si corresponde)\n",
    "* **barrio**: barrio según cartografía oficial\n",
    "* **properati_url**: url de la publicación en Properati\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Importando las bibliotecas necesarias\n",
    "\n",
    "Para poder trabajar en la limpieza de datos y la visualización de nuestro dataset vamos a utilizar las bibliotecas numpy, pandas, matplotlib.pylab y seaborn. \n",
    "\n",
    "** 1) Importá ** cada una de ellas con los siguientes nombres: \n",
    "- **numpy** como **np**\n",
    "- **pandas** como **pd**\n",
    "- **matplotlib.pylab** como **plt**\n",
    "- **seaborn** como **sns**\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importar bibliotecas en esta celda\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pylab as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# La siguiente línea es para ver las imagenes dentro del notebook\n",
    "%matplotlib inline\n",
    "# Acá configuramos el tamaño de las figuras\n",
    "plt.rcParams['figure.figsize'] = (12,8)\n",
    "# Seteamos opciones de pandas sobre las columnas y su ancho\n",
    "pd.set_option('max_columns', 120)\n",
    "pd.set_option('max_colwidth', 5000)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cargando dataset de propiedades\n",
    "\n",
    "**2) Cargá** el dataset que se encuentra en la carpeta *dataset/* y guardalo en una variable llamada **df**. \n",
    "No te olvides de utilizar **parse_dates** sobre la columna **created_on**\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# cargá el dataset en esta celda\n",
    "df = pd.read_csv('datos_properati.csv', parse_dates = ['created_on'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploración de datos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lo primero que vamos a hacer es ver qué forma tiene el dataset que acabamos de cargar. Para eso vamos a mostrar las primeras 3 entradas del mismo y utilizar la propiedad `shape` para ver la cantidad de instancias y atributos con los que vamos a estar trabajando."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(18979, 17)\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>created_on</th>\n",
       "      <th>operation</th>\n",
       "      <th>property_type</th>\n",
       "      <th>place_with_parent_names</th>\n",
       "      <th>lat-lon</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>price_aprox_usd</th>\n",
       "      <th>surface_total_in_m2</th>\n",
       "      <th>surface_covered_in_m2</th>\n",
       "      <th>price_usd_per_m2</th>\n",
       "      <th>floor</th>\n",
       "      <th>rooms</th>\n",
       "      <th>expenses</th>\n",
       "      <th>properati_url</th>\n",
       "      <th>barrio</th>\n",
       "      <th>comuna</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-09-22</td>\n",
       "      <td>sell</td>\n",
       "      <td>house</td>\n",
       "      <td>|Argentina|Capital Federal|Palermo|Palermo Viejo|</td>\n",
       "      <td>-34.5889531469,-58.4298817776</td>\n",
       "      <td>-34.588953</td>\n",
       "      <td>-58.429882</td>\n",
       "      <td>2000000.0</td>\n",
       "      <td>312.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6410.25641</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://www.properati.com.ar/2os_venta_casa_palermo-viejo_rent-ba</td>\n",
       "      <td>PALERMO</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-08-01</td>\n",
       "      <td>sell</td>\n",
       "      <td>store</td>\n",
       "      <td>|Argentina|Capital Federal|Palermo|</td>\n",
       "      <td>-34.586069,-58.414431</td>\n",
       "      <td>-34.586069</td>\n",
       "      <td>-58.414431</td>\n",
       "      <td>790000.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>6528.92562</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://www.properati.com.ar/1cl7l_venta_local_palermo_santa-fe-av-al-3500_sotano_garage_subte-linea-d_deptos-temporarios</td>\n",
       "      <td>PALERMO</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-08-02</td>\n",
       "      <td>sell</td>\n",
       "      <td>store</td>\n",
       "      <td>|Argentina|Capital Federal|Palermo|Palermo Soho|</td>\n",
       "      <td>-34.589723,-58.4325664</td>\n",
       "      <td>-34.589723</td>\n",
       "      <td>-58.432566</td>\n",
       "      <td>343000.0</td>\n",
       "      <td>122.0</td>\n",
       "      <td>108.0</td>\n",
       "      <td>2811.47541</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://www.properati.com.ar/1cm8w_venta_local_palermo-soho_cabrera-jose-a-_5000_jardin_lavadero_patio_piscina_toilette_garage_ojo-propiedades</td>\n",
       "      <td>PALERMO</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  created_on operation property_type  \\\n",
       "0 2017-09-22      sell         house   \n",
       "1 2017-08-01      sell         store   \n",
       "2 2017-08-02      sell         store   \n",
       "\n",
       "                             place_with_parent_names  \\\n",
       "0  |Argentina|Capital Federal|Palermo|Palermo Viejo|   \n",
       "1                |Argentina|Capital Federal|Palermo|   \n",
       "2   |Argentina|Capital Federal|Palermo|Palermo Soho|   \n",
       "\n",
       "                         lat-lon        lat        lon  price_aprox_usd  \\\n",
       "0  -34.5889531469,-58.4298817776 -34.588953 -58.429882        2000000.0   \n",
       "1          -34.586069,-58.414431 -34.586069 -58.414431         790000.0   \n",
       "2         -34.589723,-58.4325664 -34.589723 -58.432566         343000.0   \n",
       "\n",
       "   surface_total_in_m2  surface_covered_in_m2  price_usd_per_m2  floor  rooms  \\\n",
       "0                312.0                    NaN        6410.25641    NaN    NaN   \n",
       "1                121.0                  121.0        6528.92562    NaN    NaN   \n",
       "2                122.0                  108.0        2811.47541    NaN    2.0   \n",
       "\n",
       "   expenses  \\\n",
       "0       NaN   \n",
       "1       NaN   \n",
       "2       NaN   \n",
       "\n",
       "                                                                                                                                    properati_url  \\\n",
       "0                                                                               https://www.properati.com.ar/2os_venta_casa_palermo-viejo_rent-ba   \n",
       "1                       https://www.properati.com.ar/1cl7l_venta_local_palermo_santa-fe-av-al-3500_sotano_garage_subte-linea-d_deptos-temporarios   \n",
       "2  https://www.properati.com.ar/1cm8w_venta_local_palermo-soho_cabrera-jose-a-_5000_jardin_lavadero_patio_piscina_toilette_garage_ojo-propiedades   \n",
       "\n",
       "    barrio  comuna  \n",
       "0  PALERMO    14.0  \n",
       "1  PALERMO    14.0  \n",
       "2  PALERMO    14.0  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df.shape)\n",
    "df.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Categorías de propiedades\n",
    "En primer lugar vamos a ver qué categorías existen en el campo `property_type`. Y de esas categorías cuantos valores tenemos de cada uno.\n",
    "\n",
    "**3) Mostrá** las categorías de `property_type` y cuántas instancias hay de cada una.\n",
    "\n",
    "Recordá utilizar funciones del **dataframe** de pandas para realizar esto."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "apartment    14807\n",
       "house         1618\n",
       "PH            1463\n",
       "store         1091\n",
       "Name: property_type, dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# mostrá las categorías del campo property_type y mostrá la cant. de valores en esta celda\n",
    "df.property_type.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Filtrando las categorías\n",
    "\n",
    "De las categorías existentes sólo queremos quedarnos con las que sean casas (house), departamentos (apartment) o PH (PH). \n",
    "\n",
    "Para esto, \n",
    "\n",
    "**4) filtrá** el `dataframe` original creando un **nuevo** `dataframe` llamado `df_filtered`. Este es el `dataframe` que seguirás usando a lo largo del análisis.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "apartment    14807\n",
       "house         1618\n",
       "PH            1463\n",
       "Name: property_type, dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Filtrá el dataframe en esta celda\n",
    "df_filtered = df[df.property_type != \"store\"]\n",
    "df_filtered.property_type.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualizando las categorías\n",
    "**5)** Utilizando `seaborn` **creá un countplot** para visualizar las categorías de `df_filtered` en el campo `property_type` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualizá las categorías en esta celda\n",
    "ax = sns.countplot(x=\"property_type\", data=df_filtered)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Valores Faltantes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vamos a querer descartar aquellas instancias que tengan valores nulos en columnas importantes para el análisis del mercado.\n",
    "Estas columnas son las que informan **superficie (total o cubierta)**, **cantidad de ambientes**, **precio** y **precio por metro cuadrado**.\n",
    "\n",
    "**6) Quitá** del `dataframe` las instancias que no tengan ningún valor en los campos nombrados."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Quitá las instancias con valores nules en esta celda\n",
    "#superficie (total o cubierta) = surface_total_in_m2 / surface_covered_in_m2\n",
    "#cantidad de ambientes = rooms\n",
    "#precio = price_aprox_usd\n",
    "#precio por metro cuadrado = price_usd_per_m2\n",
    "superficie      =  df_filtered.surface_covered_in_m2.notnull() | df_filtered.surface_total_in_m2.notnull() \n",
    "cant_ambientes  =  df_filtered.rooms.notnull()\n",
    "precio          =  df_filtered.price_aprox_usd.notnull()\n",
    "precio_x_m2     =  df_filtered.price_usd_per_m2.notnull()\n",
    "df_filtered     =  df_filtered[superficie & cant_ambientes & precio & precio_x_m2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**7)** A continuación **mostrá** cuantas instancias son las que descartaste a partir de quitar las que tenían valores nulos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INSTANCIAS DESCARTADAS : 6649 \n"
     ]
    }
   ],
   "source": [
    "# Mostrá en esta celda la cantidad de instancias descartadas\n",
    "df_instancias = df.shape[0]\n",
    "df_filtrados = df_filtered.shape[0]\n",
    "df_descartados = df_instancias - df_filtrados\n",
    "print(\"\"\"INSTANCIAS DESCARTADAS : %s \"\"\" % (df_descartados))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Posiblemente existan otros atributos con valores faltantes en el dataset, pero que no son de tanta importancia. \n",
    "\n",
    "**8) Mostrá** qué otros atributos tienen valores faltantes en el `dataframe`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Mostrá en esta celda los otros atributos con valores faltantes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Desafío: planteá alguna estrategia posible para reemplazar los valores faltantes en el atributo `surface_total_in_m2`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Datos erróneos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "El siguiente paso es analizar errores que puede llegar a haber en el dataset. \n",
    "\n",
    "Por ejemplo, la **superficie total no puede ser menor que la cubierta**. Si esto sucede, podemos optar por eliminar esas filas o intentar corregir el problema. Veamos en primer lugar si esto sucede.\n",
    "\n",
    "**9) Mostrá** cuantas instancias existen que tengan el problema de este dato erróneo y guardalas en una variable llamada `wrong_surface_rows`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CANTIDAD DE PROBLEMAS EN LA SUPERFICIE : 73 \n"
     ]
    }
   ],
   "source": [
    "# Mostrá en esta celda cuántas instancias tienen errores en la superficie\n",
    "wrong_surface_rows = df_filtered[df_filtered.surface_total_in_m2 < df_filtered.surface_covered_in_m2]\n",
    "print(\"\"\"CANTIDAD DE PROBLEMAS EN LA SUPERFICIE : %s \"\"\" % (wrong_surface_rows.shape[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Proponemos no borrar estas instancias erróneas, sino intercambiar los valores de **superficie total** con **superficie cubierta** en caso de que se cumpla la condición mencionada."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_filtered.at[wrong_surface_rows.index, 'surface_total_in_m2'] = wrong_surface_rows.surface_covered_in_m2\n",
    "df_filtered.at[wrong_surface_rows.index, 'surface_covered_in_m2'] = wrong_surface_rows.surface_total_in_m2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Por otro lado, sucede que las inmobiliarias cuando no cuentan con las coordenadas reales de las propiedades utilizan valores que no son correctos. Por ejemplo, el punto centro de cada barrio o puntos de interés como el Obelisco. Esto puede afectar el análisis relacionado con la ubicación de cada propiedad.\n",
    "\n",
    "Veamos en primer lugar si hay coordenadas que se repiten."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "lat         lon       \n",
       "-34.589546  -58.397364    149\n",
       "-34.571150  -58.423297    118\n",
       "-34.562084  -58.456686     77\n",
       "-34.620928  -58.445874     45\n",
       "-34.609821  -58.392606     36\n",
       "-34.610122  -58.405941     31\n",
       "-34.597679  -58.443019     31\n",
       "-34.621184  -58.373116     31\n",
       "-34.617719  -58.362056     31\n",
       "-34.598647  -58.439018     29\n",
       "dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "conteo_puntos = df_filtered.groupby(['lat', 'lon']).size()\n",
    "conteo_puntos[conteo_puntos > 1].sort_values(ascending=False).head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Por ejemplo, para el caso del tercer punto, se puede verificar que corresponde a la esquina de Av. Cabildo y Juramento.\n",
    "Nos quedamos entonces con las propiedades con coordenadas que se repitan hasta 5 veces."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "conteo_puntos.name = 'conteo'\n",
    "\n",
    "df_filtered = df_filtered.join(conteo_puntos, on=['lat', 'lon'])\n",
    "\n",
    "df_filtered = df_filtered[df_filtered.conteo <= 5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Eliminamos el atributo `conteo` dado que ya no nos sirve."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_filtered.drop('conteo', inplace=True, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Estudiando los barrios"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Barrios con más edificios y barrios con más casas\n",
    "Al saber qué barrios son los que tienen más cantidad de edificios en venta y cuáles son los que tienen más casas podemos darnos una idea de qué tipo de barrio es. Si es más residencial, con más población, poder adquisitivo, etc. \n",
    "Para eso vamos a empezar calculando los porcentajes de cada tipo de propiedad según el barrio."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "total_propiedades = df_filtered.groupby('barrio')['property_type'].count().values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "porcentaje_casas_ph = df_filtered[df_filtered.property_type.isin(['PH', 'house'])].groupby('barrio').count().property_type/total_propiedades\n",
    "porcentaje_edificios = df_filtered[df_filtered.property_type == 'apartment'].groupby('barrio').count().property_type/total_propiedades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Barrios con mayor porcentaje de edificios: \n",
      " barrio\n",
      "PUERTO MADERO    0.995413\n",
      "RETIRO           0.992736\n",
      "SAN NICOLAS      0.989691\n",
      "RECOLETA         0.986654\n",
      "BELGRANO         0.951760\n",
      "Name: property_type, dtype: float64\n",
      "Barrios con mayor porcentaje de casas y phs: \n",
      "  barrio\n",
      "VERSALLES            0.608696\n",
      "MATADEROS            0.604651\n",
      "VILLA RIACHUELO      0.555556\n",
      "PARQUE AVELLANEDA    0.544118\n",
      "VILLA REAL           0.500000\n",
      "Name: property_type, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print(\"Barrios con mayor porcentaje de edificios: \\n\", porcentaje_edificios.sort_values()[::-1][:5])\n",
    "print(\"Barrios con mayor porcentaje de casas y phs: \\n \", porcentaje_casas_ph.sort_values()[::-1][:5])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para darnos una idea de las diferencias entre los porcentajes de edificios y casas en estos barrios, vamos a graficarlos\n",
    "en forma de torta."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x214c1bba828>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "barrios_casas = porcentaje_casas_ph.sort_values()[::-1][:5].index\n",
    "barrios_edificios = porcentaje_edificios.sort_values()[::-1][:5].index\n",
    "fig, axs = plt.subplots(1,2,figsize=(14,7))\n",
    "df_filtered[df_filtered.barrio.isin(barrios_edificios)].property_type.value_counts().plot(x=None,y=None, kind='pie', startangle=30, ax=axs[0],autopct='%1.2f%%')\n",
    "df_filtered[df_filtered.barrio.isin(barrios_casas)].property_type.value_counts().plot(x=None,y=None, kind='pie', ax=axs[1],autopct='%1.2f%%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Barrios más caros y más baratos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Veamos primero la distribución del precio por metro cuadrado para todo el dataset. De esta manera, vamos a remover valores que consideremos extremos.\n",
    "\n",
    "En el siguiente proyecto vamos a volver sobre este tema con más profundidad.\n",
    "\n",
    "**10)** Tomando el atributo `price_usd_per_m2` **calculá** su **mínimo**, **máximo**, **promedio (mean)**, **mediana (median)** y **desvío estándar (std)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "DISTRIBUCION DEL PRECIO POR METRO CUADRADO:\n",
      "    minimo            : $0.96\n",
      "    maximo            : $142450.14\n",
      "    media             : $2784.33\n",
      "    mediana           : $2588.24\n",
      "    desviacion_tipica : $1977.41\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Mostrá en esta celda los cálculos\n",
    "\n",
    "minimo = df_filtered[\"price_usd_per_m2\"].min()\n",
    "maximo = df_filtered[\"price_usd_per_m2\"].max()\n",
    "media = df_filtered[\"price_usd_per_m2\"].mean()\n",
    "mediana = df_filtered[\"price_usd_per_m2\"].median()\n",
    "desviacion_tipica = df_filtered[\"price_usd_per_m2\"].std(ddof=0)\n",
    "print(\"\"\"\n",
    "DISTRIBUCION DEL PRECIO POR METRO CUADRADO:\n",
    "    minimo            : $%.2f\n",
    "    maximo            : $%.2f\n",
    "    media             : $%.2f\n",
    "    mediana           : $%.2f\n",
    "    desviacion_tipica : $%.2f\n",
    "\"\"\" % (minimo,maximo,media, mediana, desviacion_tipica))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "También podemos calcular otros estadísticos como los cuartiles. \n",
    "\n",
    "**11) Mostrá** el **primer y tercer cuartil** y luego también el **rango intercuartilíco**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "DISTRIBUCION DEL PRECIO POR METRO CUADRADO:\n",
      "    primer cuartil            : $2036.73\n",
      "    tercer cuartil            : $3205.13\n",
      "    rango intercuartilíco     : $1168.39\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Mostrá los cuartiles en esta celda\n",
    "primer_cuartil = df_filtered[\"price_usd_per_m2\"].quantile(0.25)\n",
    "tercer_cuartil = df_filtered[\"price_usd_per_m2\"].quantile(0.75)\n",
    "rango_intercuartilíco = tercer_cuartil - primer_cuartil\n",
    "print(\"\"\"\n",
    "DISTRIBUCION DEL PRECIO POR METRO CUADRADO:\n",
    "    primer cuartil            : $%.2f\n",
    "    tercer cuartil            : $%.2f\n",
    "    rango intercuartilíco     : $%.2f\n",
    "\"\"\" % (primer_cuartil,tercer_cuartil, rango_intercuartilíco))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Quitando valores extremos\n",
    "Eliminemos aquellas observaciones cuyo precio por metro cuadrado este fuera de los percentiles 10 y 90. De este modo intentaremos remover valores extremos de una manera sencilla para tener un análisis más robusto."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "p10 = np.percentile(df_filtered.price_usd_per_m2, 10)\n",
    "p90 = np.percentile(df_filtered.price_usd_per_m2, 90)\n",
    "df_filtered = df_filtered[(df_filtered.price_usd_per_m2 >= p10) & (df_filtered.price_usd_per_m2 <= p90)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A partir de quitar estos valores van a cambiar los estadísticos que calculamos.\n",
    "\n",
    "**12) Volvé a calcular** los mismos estadísticos (min, max, promedio, etc) para ver los cambios.\n",
    "¿Por qué será que la **mediana** no cambia?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "DISTRIBUCION DEL PRECIO POR METRO CUADRADO:\n",
      "    minimo            : $1552.63\n",
      "    maximo            : $4095.89\n",
      "    media             : $2632.26\n",
      "    mediana           : $2588.76\n",
      "    desviacion_tipica : $608.25\n",
      "\n",
      "\n",
      "    primer cuartil            : $2150.00\n",
      "    tercer cuartil            : $3056.75\n",
      "    rango intercuartilíco     : $906.75\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Realizá los calculos en esta celda\n",
    "minimo = df_filtered[\"price_usd_per_m2\"].min()\n",
    "maximo = df_filtered[\"price_usd_per_m2\"].max()\n",
    "media = df_filtered[\"price_usd_per_m2\"].mean()\n",
    "mediana = df_filtered[\"price_usd_per_m2\"].median()\n",
    "desviacion_tipica = df_filtered[\"price_usd_per_m2\"].std(ddof=0)\n",
    "print(\"\"\"\n",
    "DISTRIBUCION DEL PRECIO POR METRO CUADRADO:\n",
    "    minimo            : $%.2f\n",
    "    maximo            : $%.2f\n",
    "    media             : $%.2f\n",
    "    mediana           : $%.2f\n",
    "    desviacion_tipica : $%.2f\n",
    "\"\"\" % (minimo,maximo,media, mediana, desviacion_tipica))\n",
    "\n",
    "primer_cuartil = df_filtered[\"price_usd_per_m2\"].quantile(0.25)\n",
    "tercer_cuartil = df_filtered[\"price_usd_per_m2\"].quantile(0.75)\n",
    "rango_intercuartilíco = tercer_cuartil - primer_cuartil\n",
    "print(\"\"\"\n",
    "    primer cuartil            : $%.2f\n",
    "    tercer cuartil            : $%.2f\n",
    "    rango intercuartilíco     : $%.2f\n",
    "\"\"\" % (primer_cuartil,tercer_cuartil, rango_intercuartilíco))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vamos a graficar ahora los barrios según el precio por metro cuadrado."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Anaconda3\\envs\\Data Science\\lib\\site-packages\\scipy\\stats\\stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.xticks(rotation = 90)\n",
    "barrios = df_filtered.groupby('barrio').mean().sort_values('price_usd_per_m2', ascending=False).index\n",
    "sns.barplot(x=\"barrio\", y=\"price_usd_per_m2\", order=barrios, data=df_filtered);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Más caros"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "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>price_usd_per_m2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>barrio</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PUERTO MADERO</th>\n",
       "      <td>6156.633183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PALERMO</th>\n",
       "      <td>3632.057960</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RETIRO</th>\n",
       "      <td>3417.635657</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RECOLETA</th>\n",
       "      <td>3346.748987</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BELGRANO</th>\n",
       "      <td>3338.725206</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               price_usd_per_m2\n",
       "barrio                         \n",
       "PUERTO MADERO       6156.633183\n",
       "PALERMO             3632.057960\n",
       "RETIRO              3417.635657\n",
       "RECOLETA            3346.748987\n",
       "BELGRANO            3338.725206"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "table = pd.pivot_table(df, values='price_usd_per_m2', index=['barrio'], aggfunc=np.mean)\n",
    "mas_caros = table.sort_values('price_usd_per_m2', ascending=False).iloc[:5]\n",
    "mas_caros"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Más baratos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "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>price_usd_per_m2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>barrio</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>VILLA SOLDATI</th>\n",
       "      <td>920.825753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VILLA LUGANO</th>\n",
       "      <td>1179.846071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VILLA RIACHUELO</th>\n",
       "      <td>1274.809950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PARQUE AVELLANEDA</th>\n",
       "      <td>1386.260577</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUEVA POMPEYA</th>\n",
       "      <td>1539.235869</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   price_usd_per_m2\n",
       "barrio                             \n",
       "VILLA SOLDATI            920.825753\n",
       "VILLA LUGANO            1179.846071\n",
       "VILLA RIACHUELO         1274.809950\n",
       "PARQUE AVELLANEDA       1386.260577\n",
       "NUEVA POMPEYA           1539.235869"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "table = pd.pivot_table(df, values='price_usd_per_m2', index=['barrio'], aggfunc=np.mean)\n",
    "mas_baratos = table.sort_values('price_usd_per_m2', ascending=True).iloc[:5]\n",
    "mas_baratos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Graficando propiedades de los barrios\n",
    "\n",
    "Ahora que ya sabemos cuales son los barrios más caros y cuales son los más baratos vamos a ver algunas comparaciones gráficas para entender mejor sus diferencias en diferentes atributos del dataset. Vamos a ver las distribuciones según:\n",
    "- Superficie total en m2\n",
    "- Cuartos totales \n",
    "- Precio en dolares por m2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "barrios_caros = mas_caros.index \n",
    "barrios_baratos = mas_baratos.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Anaconda3\\envs\\Data Science\\lib\\site-packages\\scipy\\stats\\stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x214c264cfd0>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.distplot(df_filtered[df_filtered.barrio.isin(barrios_baratos)].surface_total_in_m2, label=\"Barrios baratos\")\n",
    "sns.distplot(df_filtered[df_filtered.barrio.isin(barrios_caros)].surface_total_in_m2, label=\"Barrios caros\")\n",
    "g.set_xlim(0, 400)\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Anaconda3\\envs\\Data Science\\lib\\site-packages\\scipy\\stats\\stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x214c24c9278>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.distplot(df_filtered[df_filtered.barrio.isin(barrios_baratos)].rooms, label=\"Barrios baratos\")\n",
    "sns.distplot(df_filtered[df_filtered.barrio.isin(barrios_caros)].rooms, label=\"Barrios caros\")\n",
    "g.set_xlim(0, 10)\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Anaconda3\\envs\\Data Science\\lib\\site-packages\\scipy\\stats\\stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x214c2565160>"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.distplot(df_filtered[df_filtered.barrio.isin(barrios_baratos)].price_usd_per_m2, label=\"Barrios baratos\")\n",
    "sns.distplot(df_filtered[df_filtered.barrio.isin(barrios_caros)].price_usd_per_m2, label=\"Barrios caros\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Barrios que quieren ser otros"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Basado en información de mercado, desde el portal Properati han señalado que las inmobiliarias suelen publicar las propiedades ubicadas en barrios que no necesariamente se corresponden con la ubicación geográfica real. Según mencionan en la nota ([blog)](http://blog.properati.com.ar/barrios-que-quieren-ser-otros/), esto corresponde a una lógica comercial en el corredor norte y a usos y costumbres en zona oeste.\n",
    "\n",
    "Analicemos qué sucede para nuestro dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dada la estructura del atributo `place_with_parent_names`, en primer lugar vamos a extraer el campo asociado al barrio publicado para poder comparar con el que surge de la cartografía oficial. Esto lo vamos a agregar en una nueva columna llamada `barrio_publicado`. \n",
    "\n",
    "De este modo, lo que vamos a analizar es como se relacionan las columnas `barrio` y `barrio publicado`. Si efectivamente hay barrios que quieren ser otros, esperaríamos ver propiedades ubicadas en geográficamente en un barrio distinto al que dice la publicación. Por ejemplo, si una propiedad está ubicada (dadas sus coordenadas) en Villa Crespo pero en la publicación dice Palermo, entonces decimos que hay indicios que Villa Crespo quiere ser Palermo.\n",
    "\n",
    "Viendo la estructura de este atributo, podemos notar que se compone por una sequencia de lugares separados por `|`. Por ejemplo, en `|Argentina|Capital Federal|Palermo|` vemos que primero está el país, luego la ciudad, luego el barrio y en algunos casos hay una jerarquía más, relacionado con zonas dentro del barrio (como Palermo Soho dentro de Palermo). \n",
    "\n",
    "De este modo, una estrategia posible para resolver esto es separar las jerarquías por `|` y extraer el elemento asociado al barrio. Por ejemplo, para `|Argentina|Capital Federal|Palermo|` queda `['', 'Argentina', 'Capital Federal', 'Palermo', '']`. Extrayendo el tercer elemento obtendremos el barrio independientemente de si hay lugares luego en la jerarquía. Aquellos que no tengan barrio no serán considerados puesto que estarían ubicados en Capital Federal y no sería informativo.\n",
    "\n",
    "Lo pasamos a mayúsculas y quitamos las tildes para poder comparar con el barrio oficial."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5, 6, 4], dtype=int64)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_filtered['place_with_parent_names'].apply(lambda x: len(x.split(\"|\"))).unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como algunos tienen 4 elementos, debemos quitarlos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "split_place_filter = df_filtered['place_with_parent_names'].apply(lambda x: len(x.split(\"|\"))!=4)\n",
    "df_filtered = df_filtered.loc[split_place_filter]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_filtered['barrio_publicado'] = df_filtered['place_with_parent_names'].apply( lambda x: x.split(\"|\")[3].upper())\n",
    "df_filtered['barrio_publicado'] = df_filtered['barrio_publicado'].str.normalize('NFKD').str.encode('ascii', errors='ignore').str.decode('utf-8')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Comparemos entonces el barrio publicado con el real. Nos vamos a quedar con aquellas propiedades donde nos coincidan para luego hacer el conteo por barrio."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "barrios_match = (df_filtered['barrio_publicado'] != df_filtered['barrio'])\n",
    "df_filtered_barrio_not_match = df_filtered.loc[barrios_match]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Con una tabla de doble entrada podemos identificar para cada barrio cuál es al que más quiere parecerse."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "barrio\n",
       "AGRONOMIA                VILLA PUEYRREDON\n",
       "ALMAGRO                         CABALLITO\n",
       "BALVANERA                            ONCE\n",
       "BARRACAS                        SAN TELMO\n",
       "BELGRANO                            NUNEZ\n",
       "BOCA                             BARRACAS\n",
       "BOEDO                             ALMAGRO\n",
       "CABALLITO               PARQUE CENTENARIO\n",
       "CHACARITA                      COLEGIALES\n",
       "COGHLAN                          BELGRANO\n",
       "COLEGIALES                       BELGRANO\n",
       "CONSTITUCION                    SAN TELMO\n",
       "FLORES                          CABALLITO\n",
       "FLORESTA                           FLORES\n",
       "LINIERS                         VERSALLES\n",
       "MATADEROS                         LINIERS\n",
       "MONSERRAT                       SAN TELMO\n",
       "MONTE CASTRO                 VILLA DEVOTO\n",
       "NUEVA POMPEYA            PARQUE PATRICIOS\n",
       "NUEZ                                NUNEZ\n",
       "PALERMO                          BELGRANO\n",
       "PARQUE AVELLANEDA                FLORESTA\n",
       "PARQUE CHACABUCO                CABALLITO\n",
       "PARQUE CHAS                 VILLA URQUIZA\n",
       "PARQUE PATRICIOS                    BOEDO\n",
       "PATERNAL                        AGRONOMIA\n",
       "RECOLETA                     BARRIO NORTE\n",
       "RETIRO                           RECOLETA\n",
       "SAAVEDRA                            NUNEZ\n",
       "SAN CRISTOBAL                       BOEDO\n",
       "SAN NICOLAS          CENTRO / MICROCENTRO\n",
       "SAN TELMO                        BARRACAS\n",
       "VELEZ SARSFIELD                  FLORESTA\n",
       "VILLA CRESPO                      PALERMO\n",
       "VILLA DEL PARQUE             VILLA DEVOTO\n",
       "VILLA DEVOTO             VILLA DEL PARQUE\n",
       "VILLA GRAL. MITRE     VILLA GENERAL MITRE\n",
       "VILLA LURO                      VERSALLES\n",
       "VILLA ORTUZAR               VILLA URQUIZA\n",
       "VILLA PUEYRREDON             VILLA DEVOTO\n",
       "VILLA REAL                      VERSALLES\n",
       "VILLA RIACHUELO              VILLA LUGANO\n",
       "VILLA SANTA RITA         VILLA DEL PARQUE\n",
       "VILLA SOLDATI                VILLA LUGANO\n",
       "VILLA URQUIZA                    BELGRANO\n",
       "dtype: object"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "table = pd.crosstab(df_filtered_barrio_not_match.barrio, df_filtered_barrio_not_match.barrio_publicado)\n",
    "table.idxmax(axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podemos ver entonces que los barrios tienden a querer parecerse a barrios con precio con metro cuadrado más caro. Por ejemplo Villa Crespo con Palermo, Constitución con San Telmo y Villa Urquiza con Belgrano. \n",
    "\n",
    "Por otro lado, vemos también que los barrios suelen querer parecerse a sus barrios vecinos."
   ]
  }
 ],
 "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": "3.5.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}