KNN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "34d4720c",
   "metadata": {},
   "source": [
    "# Algoritmo k-Nearest Neighbors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "138a2117",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.colors import ListedColormap\n",
    "import matplotlib.patches as mpatches\n",
    "import seaborn as sb\n",
    " \n",
    "%matplotlib inline\n",
    "plt.rcParams['figure.figsize'] = (16, 9)\n",
    "plt.style.use('ggplot')\n",
    " \n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import confusion_matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "652a588c",
   "metadata": {},
   "source": [
    "Para nuestro ejercicio tomaremos 257 registros con Opiniones de usuarios sobre una app (Reviews). Utilizaremos 2 columnas de datos como fuente de alimento del algoritmo. Recordad que sólo tomaré 2 features para poder graficar en 2 dimensiones, PERO para un problema “en la vida real” conviene tomar más características de lo que sea que queramos resolver. Esto es únicamente con fines de enseñanza. \n",
    "\n",
    "Las columnas que utilizaremos serán: wordcount con la cantidad de palabras utilizadas y sentimentValue con un valor entre -4 y 4 que indica si el comentario fue valorado como positivo o negativo. Nuestras etiquetas, serán las estrellas que dieron los usuarios a la app, que son valores discretos del 1 al 5.\n",
    "\n",
    "Podemos pensar que si el usuario puntúa con más estrellas, tendrá un sentimiento positivo, pero no necesariamente siempre es así."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "37644937",
   "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>Review Title</th>\n",
       "      <th>Review Text</th>\n",
       "      <th>wordcount</th>\n",
       "      <th>titleSentiment</th>\n",
       "      <th>textSentiment</th>\n",
       "      <th>Star Rating</th>\n",
       "      <th>sentimentValue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Sin conexión</td>\n",
       "      <td>Hola desde hace algo más de un mes me pone sin...</td>\n",
       "      <td>23</td>\n",
       "      <td>negative</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.486389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>faltan cosas</td>\n",
       "      <td>Han mejorado la apariencia pero no</td>\n",
       "      <td>20</td>\n",
       "      <td>negative</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.586187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Es muy buena lo recomiendo</td>\n",
       "      <td>Andres e puto amoooo</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.602240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Version antigua</td>\n",
       "      <td>Me gustana mas la version anterior esta es mas...</td>\n",
       "      <td>17</td>\n",
       "      <td>NaN</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.616271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Esta bien</td>\n",
       "      <td>Sin ser la biblia.... Esta bien</td>\n",
       "      <td>6</td>\n",
       "      <td>negative</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.651784</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Buena</td>\n",
       "      <td>Nada del otro mundo pero han mejorado mucho</td>\n",
       "      <td>8</td>\n",
       "      <td>positive</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.720443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>De gran ayuda</td>\n",
       "      <td>Lo malo q necesita de …,pero la app es muy buena</td>\n",
       "      <td>23</td>\n",
       "      <td>positive</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.726825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Muy buena</td>\n",
       "      <td>Estaba más acostumbrado al otro diseño, pero e...</td>\n",
       "      <td>16</td>\n",
       "      <td>positive</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.736769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Ta to guapa.</td>\n",
       "      <td>Va de escándalo</td>\n",
       "      <td>21</td>\n",
       "      <td>positive</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.765284</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Se han corregido</td>\n",
       "      <td>Han corregido muchos fallos pero el diseño es ...</td>\n",
       "      <td>13</td>\n",
       "      <td>negative</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.797961</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Review Title  \\\n",
       "0                Sin conexión   \n",
       "1                faltan cosas   \n",
       "2  Es muy buena lo recomiendo   \n",
       "3             Version antigua   \n",
       "4                   Esta bien   \n",
       "5                       Buena   \n",
       "6               De gran ayuda   \n",
       "7                   Muy buena   \n",
       "8                Ta to guapa.   \n",
       "9            Se han corregido   \n",
       "\n",
       "                                         Review Text  wordcount  \\\n",
       "0  Hola desde hace algo más de un mes me pone sin...         23   \n",
       "1                 Han mejorado la apariencia pero no         20   \n",
       "2                               Andres e puto amoooo          4   \n",
       "3  Me gustana mas la version anterior esta es mas...         17   \n",
       "4                    Sin ser la biblia.... Esta bien          6   \n",
       "5        Nada del otro mundo pero han mejorado mucho          8   \n",
       "6   Lo malo q necesita de …,pero la app es muy buena         23   \n",
       "7  Estaba más acostumbrado al otro diseño, pero e...         16   \n",
       "8                                    Va de escándalo         21   \n",
       "9  Han corregido muchos fallos pero el diseño es ...         13   \n",
       "\n",
       "  titleSentiment textSentiment  Star Rating  sentimentValue  \n",
       "0       negative      negative            1       -0.486389  \n",
       "1       negative      negative            1       -0.586187  \n",
       "2            NaN      negative            1       -0.602240  \n",
       "3            NaN      negative            1       -0.616271  \n",
       "4       negative      negative            1       -0.651784  \n",
       "5       positive      negative            1       -0.720443  \n",
       "6       positive      negative            1       -0.726825  \n",
       "7       positive      negative            1       -0.736769  \n",
       "8       positive      negative            1       -0.765284  \n",
       "9       negative      negative            1       -0.797961  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dt = pd.read_csv(\"data/reviews_sentiment.csv\",sep=';')\n",
    "dt.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fc5ab9ad",
   "metadata": {},
   "source": [
    "Aprovechamos para realizar un breve resumen de los datos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "802dd67f",
   "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>wordcount</th>\n",
       "      <th>Star Rating</th>\n",
       "      <th>sentimentValue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>257.000000</td>\n",
       "      <td>257.000000</td>\n",
       "      <td>257.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>11.501946</td>\n",
       "      <td>3.420233</td>\n",
       "      <td>0.383849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>13.159812</td>\n",
       "      <td>1.409531</td>\n",
       "      <td>0.897987</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-2.276469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>-0.108144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>7.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.264091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>16.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>0.808384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>103.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>3.264579</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        wordcount  Star Rating  sentimentValue\n",
       "count  257.000000   257.000000      257.000000\n",
       "mean    11.501946     3.420233        0.383849\n",
       "std     13.159812     1.409531        0.897987\n",
       "min      1.000000     1.000000       -2.276469\n",
       "25%      3.000000     3.000000       -0.108144\n",
       "50%      7.000000     3.000000        0.264091\n",
       "75%     16.000000     5.000000        0.808384\n",
       "max    103.000000     5.000000        3.264579"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dt.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b35d1925",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Review Title       object\n",
       "Review Text        object\n",
       "wordcount           int64\n",
       "titleSentiment     object\n",
       "textSentiment      object\n",
       "Star Rating         int64\n",
       "sentimentValue    float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dt.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "90bd9f40",
   "metadata": {},
   "source": [
    "Son 257 registros. Las estrellas lógicamente vemos que van del 1 al 5. La cantidad de palabras van de 1 sóla hasta 103. y las valoraciones de sentimiento están entre -2.27 y 3.26 con una media de 0,38 y a partir del desvío estándar podemos ver que la mayoría están entre 0,38-0,89 y 0,38+0,89.\n",
    "\n",
    "Al ejecutar describe() solo nos describe las variables numericas"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f26a89b3",
   "metadata": {},
   "source": [
    "### Balance de datos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ec2b7df5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dt.hist()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e222ede7",
   "metadata": {},
   "source": [
    "Vemos que la distribución de “estrellas” no está balanceada… esto no es bueno. Convendría tener las mismas cantidades en las salidas, para no tener resultados “tendenciosos”. Para este ejercicio lo dejaremos así, pero en la vida real, debemos equilibrarlos. La gráfica de Valores de Sentimientos parece bastante una campana movida levemente hacia la derecha del cero y la cantidad de palabras se centra sobre todo de 0 a 10. Veamos cual es la distribución entre las etiquetas que tenemos para las estrellas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5c3e3f28",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Star Rating\n",
      "1    37\n",
      "2    24\n",
      "3    78\n",
      "4    30\n",
      "5    88\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(dt.groupby('Star Rating').size())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "378eadfc",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\vicen\\anaconda3\\lib\\site-packages\\seaborn\\categorical.py:3714: UserWarning: The `factorplot` function has been renamed to `catplot`. The original name will be removed in a future release. Please update your code. Note that the default `kind` in `factorplot` (`'point'`) has changed `'strip'` in `catplot`.\n",
      "  warnings.warn(msg)\n",
      "C:\\Users\\vicen\\anaconda3\\lib\\site-packages\\seaborn\\_decorators.py:36: FutureWarning: Pass the following variable as a keyword arg: x. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x15f16c40b80>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABEEAAAFxCAYAAACcI7beAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAdcElEQVR4nO3df5DU9WH/8dfdIQgSjoPD6nHalgAxKtFRGdBoUblMWpM0jFOdkQFD0EzFKFPNODJxqm1+dC4NhAwVtdOvgGPS2lgn15imk+nVCY76bXtG8AdSxKmNGq1wcEIw4Hnsff/IyDdUkhx4t8vxfjz+Yvdud184fObg6eezW9ff398fAAAAgGNcfa0HAAAAAFSDCAIAAAAUQQQBAAAAiiCCAAAAAEUQQQAAAIAijKj1gCPx2muv1XoCAAAAcJRqaWk55P3OBAEAAACKIIIAAAAARRBBAAAAgCKIIAAAAEARRBAAAACgCCIIAAAAUAQRBAAAACiCCAIAAAAUQQQBAAAAiiCCAAAAAEUQQQAAAIAiiCAAAABAEUQQAAAAoAgiCAAAAFAEEQQAAAAowohaDwAAAKA66jevqPUEeI/Kh79QtddyJggAAABQBBEEAAAAKIIIAgAAABRBBAEAAACKIIIAAAAARRBBAAAAgCKIIAAAAEARRBAAAACgCCIIAAAAUAQRBAAAACiCCAIAAAAUQQQBAAAAiiCCAAAAAEUQQQAAAIAiiCAAAABAEUQQAAAAoAgiCAAAAFAEEQQAAAAogggCAAAAFEEEAQAAAIogggAAAABFEEEAAACAIoggAAAAQBFGVOuFvv/97+eRRx5JXV1dTjnllFx//fXp7e3NypUrs3379kyaNCk33XRTxo4dW61JAAAAQEGqcibIzp0788///M9pb2/PihUrUqlU8sQTT6SjoyMzZszIqlWrMmPGjHR0dFRjDgAAAFCgql0OU6lU0tvbm/3796e3tzdNTU3p6urKnDlzkiRz5sxJV1dXteYAAAAAhanK5TATJkzIpz71qSxZsiQjR47MWWedlbPOOiu7du1KU1NTkqSpqSm7d++uxhwAAACgQFWJIHv27ElXV1dWr16dMWPG5Bvf+EYeffTRAT++s7MznZ2dSZL29vY0NzcP1VQAAIBj1s5aD4BDqOa/8asSQZ599tmceOKJGTduXJJk1qxZeeGFF9LY2Jienp40NTWlp6fnwNf/t7a2trS1tR243d3dXY3ZAAAAxxQfD8rRaCj+jd/S0nLI+6tyDDQ3N2fr1q15++2309/fn2effTaTJ0/Oeeedl/Xr1ydJ1q9fn5kzZ1ZjDgAAAFCgqpwJMm3atMyePTu33nprGhoa8ju/8ztpa2vLvn37snLlyjzyyCNpbm7OzTffXI05AAAAQIHq+vv7+2s94nC99tprtZ4AAAAw7NRvXlHrCfAelQ9/YdCfs6aXwwAAAADUWlUuhwEADvbD771e6wnwHh//w5NrPQEAhpQzQQAAAIAiiCAAAABAEUQQAAAAoAgiCAAAAFAEEQQAAAAogggCAAAAFEEEAQAAAIogggAAAABFEEEAAACAIoggAAAAQBFEEAAAAKAIIggAAABQBBEEAAAAKIIIAgAAABRBBAEAAACKIIIAAAAARRBBAAAAgCKIIAAAAEARRBAAAACgCCIIAAAAUAQRBAAAACiCCAIAAAAUQQQBAAAAiiCCAAAAAEUQQQAAAIAiiCAAAABAEUQQAAAAoAgiCAAAAFAEEQQAAAAogggCAAAAFEEEAQAAAIogggAAAABFEEEAAACAIoggAAAAQBFEEAAAAKAIIggAAABQBBEEAAAAKIIIAgAAABRBBAEAAACKIIIAAAAARRBBAAAAgCKIIAAAAEARRBAAAACgCCIIAAAAUAQRBAAAACiCCAIAAAAUQQQBAAAAiiCCAAAAAEUQQQAAAIAiiCAAAABAEUQQAAAAoAgiCAAAAFAEEQQAAAAogggCAAAAFEEEAQAAAIogggAAAABFEEEAAACAIoggAAAAQBFEEAAAAKAII6r1Qm+99VbuueeevPLKK6mrq8uSJUvS0tKSlStXZvv27Zk0aVJuuummjB07tlqTAAAAgIJULYKsXbs2Z599dr7whS+kr68vb7/9dr773e9mxowZmTdvXjo6OtLR0ZEFCxZUaxIAAABQkKpcDvPzn/88mzdvzqWXXpokGTFiRE444YR0dXVlzpw5SZI5c+akq6urGnMAAACAAlXlTJBt27Zl3Lhxueuuu/KTn/wkU6ZMyaJFi7Jr1640NTUlSZqamrJ79+5DPr6zszOdnZ1Jkvb29jQ3N1djNgAModdrPQDew9+x4Ni3s9YD4BCq+fOnKhFk//79eemll7J48eJMmzYta9euTUdHx4Af39bWlra2tgO3u7u7h2AlAEDZ/B0Ljn0+GYOj0VD8/GlpaTnk/VU5BiZOnJiJEydm2rRpSZLZs2fnpZdeSmNjY3p6epIkPT09GTduXDXmAAAAAAWqSgQZP358Jk6cmNdeey1J8uyzz6a1tTXnnXde1q9fnyRZv359Zs6cWY05AAAAQIGq9ukwixcvzqpVq9LX15cTTzwx119/ffr7+7Ny5co88sgjaW5uzs0331ytOQAAAEBh6vr7+/trPeJwvXtGCQAMVz/8njdG5ejz8T88udYTgCFWv3lFrSfAe1Q+/IVBf86avicIAAAAQK2JIAAAAEARRBAAAACgCCIIAAAAUAQRBAAAACiCCAIAAAAUQQQBAAAAiiCCAAAAAEUQQQAAAIAiiCAAAABAEUQQAAAAoAgiCAAAAFAEEQQAAAAogggCAAAAFEEEAQAAAIogggAAAABFEEEAAACAIoggAAAAQBFEEAAAAKAIIggAAABQBBEEAAAAKIIIAgAAABRBBAEAAACKIIIAAAAARRBBAAAAgCKIIAAAAEARRBAAAACgCCIIAAAAUIQBR5Dvfe97h7z/+9///qCNAQAAABgqA44gDz300GHdDwAAAHA0GfGbvuG5555LklQqlQO/ftcbb7yR0aNHD80yAAAAgEH0GyPI3XffnSTp7e098Oskqaury/jx47N48eKhWwcAAAAwSH5jBFm9enWS5M4778wNN9ww5IMAAAAAhsJvjCDv+uUAUqlUDvpafb0PmQEAAACObgOOIP/1X/+Ve++9Ny+//HJ6e3sP+trf//3fD/owAAAAgME04AiyevXqnHvuuVmyZElGjRo1lJsAAAAABt2AI0h3d3euuuqq1NXVDeUeAAAAgCEx4DfzmDlzZp5++umh3AIAAAAwZAZ8Jsg777yT5cuX57TTTsv48eMP+ppPjQEAAACOdgOOIK2trWltbR3KLQAAAABDZsAR5IorrhjKHQAAAABDasAR5LnnnvuVXzvzzDMHZQwAAADAUBlwBLn77rsPur179+709fVl4sSJufPOOwd9GAAAAMBgGnAEWb169UG3K5VKHnrooYwePXrQRwEAAAAMtgF/RO57Hlhfn8svvzz/+I//OJh7AAAAAIbEEUeQJHnmmWdSX/++ngIAAACgKgZ8OcySJUsOut3b25ve3t5ce+21gz4KAAAAYLANOILceOONB90eNWpUTj755IwZM2bQRwEAAAAMtgFHkNNPPz3JL94QddeuXWlsbHQpDAAAADBsDDiC7N27N/fee2+eeOKJ7N+/Pw0NDbnggguyePFiZ4MAAAAAR70Bn8qxZs2a7Nu3L8uXL8+3vvWtLF++PL29vVmzZs1Q7gMAAAAYFAOOIBs3bsyNN96YlpaWHHfccWlpacn111+fp59+eij3AQAAAAyKAUeQkSNHZvfu3Qfdt3v37owYMeAragAAAABqZsAF49JLL81XvvKVfOITn8ikSZOyffv2/NM//VPmzp07lPsAAAAABsWAI8jll1+eCRMm5LHHHsvOnTszYcKEfPrTn86ll146lPsAAAAABsWAI8jatWvz0Y9+NH/6p3964L4tW7Zk3bp1WbRo0VBsAwAAABg0A35PkMcffzwf/OAHD7pvypQpeeyxxwZ9FAAAAMBgG3AEqaurS6VSOei+SqWS/v7+QR8FAAAAMNgGHEFOO+20PPDAAwdCSKVSyYMPPpjTTjttyMYBAAAADJYBvyfIZz/72bS3t+eP//iP09zcnO7u7jQ1NeXWW28dyn0AAAAAg2LAEWTixIn52te+lhdffDE7duzIxIkTM3Xq1NTXD/hkklQqlSxbtiwTJkzIsmXLsmfPnqxcuTLbt2/PpEmTctNNN2Xs2LFH9BsBAAAA+HUGXjCS1NfXZ/r06Tn//PMzffr0wwogSfKDH/wgkydPPnC7o6MjM2bMyKpVqzJjxox0dHQc1vMBAAAADNThVYz3YceOHXnqqacyd+7cA/d1dXVlzpw5SZI5c+akq6urWnMAAACAwlQtgqxbty4LFixIXV3dgft27dqVpqamJElTU1N2795drTkAAABAYQb8niDvx49//OM0NjZmypQp2bRp02E/vrOzM52dnUmS9vb2NDc3D/ZEAKiy12s9AN7D37Hg2Lez1gPgEKr586cqEWTLli158skns2HDhvT29mbv3r1ZtWpVGhsb09PTk6ampvT09GTcuHGHfHxbW1va2toO3O7u7q7GbACAovg7Fhz7qnYpAByGofj509LScsj7qxJB5s+fn/nz5ydJNm3alIcffjhLly7N/fffn/Xr12fevHlZv359Zs6cWY05AAAAQIFqGgLnzZuXZ555JkuXLs0zzzyTefPm1XIOAAAAcAyrypkgv+yMM87IGWeckST5wAc+kNtvv73aEwAAAIACuSQMAAAAKIIIAgAAABRBBAEAAACKIIIAAAAARRBBAAAAgCJU/dNhjkav33JtrSfAe5z89f9T6wkAAADHFGeCAAAAAEUQQQAAAIAiiCAAAABAEUQQAAAAoAgiCAAAAFAEEQQAAAAogggCAAAAFEEEAQAAAIogggAAAABFEEEAAACAIoggAAAAQBFEEAAAAKAIIggAAABQBBEEAAAAKIIIAgAAABRBBAEAAACKIIIAAAAARRBBAAAAgCKIIAAAAEARRBAAAACgCCIIAAAAUAQRBAAAACiCCAIAAAAUYUStBwAAwOFYtWpVrSfAeyxdurTWE4ABcCYIAAAAUAQRBAAAACiCCAIAAAAUQQQBAAAAiiCCAAAAAEUQQQAAAIAiiCAAAABAEUQQAAAAoAgiCAAAAFAEEQQAAAAogggCAAAAFEEEAQAAAIogggAAAABFEEEAAACAIoggAAAAQBFEEAAAAKAIIggAAABQBBEEAAAAKIIIAgAAABRBBAEAAACKMKLWA4Dha9F9/7fWE+CQ1n3m/FpPAADgKORMEAAAAKAIIggAAABQBBEEAAAAKIIIAgAAABRBBAEAAACKIIIAAAAARRBBAAAAgCKIIAAAAEARRBAAAACgCCIIAAAAUIQR1XiR7u7urF69Om+++Wbq6urS1taWyy67LHv27MnKlSuzffv2TJo0KTfddFPGjh1bjUkAAABAYaoSQRoaGrJw4cJMmTIle/fuzbJly/KRj3wkP/rRjzJjxozMmzcvHR0d6ejoyIIFC6oxCQAAAChMVS6HaWpqypQpU5Iko0ePzuTJk7Nz5850dXVlzpw5SZI5c+akq6urGnMAAACAAlXlTJBftm3btrz00kuZOnVqdu3alaampiS/CCW7d+8+5GM6OzvT2dmZJGlvb09zc/Ogbnp9UJ8NBsdg/zmHkgyP48dPH44+w+PYgaPTcDl+dtZ6ABxCNY+fqkaQffv2ZcWKFVm0aFHGjBkz4Me1tbWlra3twO3u7u6hmAdHFX/O4cg5fuDIOHbgyA2X48cnY3A0Gorjp6Wl5ZD3V+0Y6Ovry4oVK3LRRRdl1qxZSZLGxsb09PQkSXp6ejJu3LhqzQEAAAAKU5UI0t/fn3vuuSeTJ0/OJz/5yQP3n3feeVm/fn2SZP369Zk5c2Y15gAAAAAFqsrlMFu2bMmjjz6aU089NbfcckuS5Kqrrsq8efOycuXKPPLII2lubs7NN99cjTkAAABAgaoSQU477bR85zvfOeTXbr/99mpMAAAAAArnfXEAAACAIoggAAAAQBFEEAAAAKAIIggAAABQBBEEAAAAKIIIAgAAABRBBAEAAACKIIIAAAAARRBBAAAAgCKIIAAAAEARRBAAAACgCCIIAAAAUAQRBAAAACiCCAIAAAAUQQQBAAAAiiCCAAAAAEUQQQAAAIAiiCAAAABAEUQQAAAAoAgiCAAAAFAEEQQAAAAogggCAAAAFEEEAQAAAIogggAAAABFEEEAAACAIoggAAAAQBFEEAAAAKAIIggAAABQBBEEAAAAKIIIAgAAABRBBAEAAACKIIIAAAAARRBBAAAAgCKIIAAAAEARRBAAAACgCCIIAAAAUAQRBAAAACiCCAIAAAAUQQQBAAAAiiCCAAAAAEUQQQAAAIAiiCAAAABAEUQQAAAAoAgiCAAAAFAEEQQAAAAogggCAAAAFEEEAQAAAIogggAAAABFEEEAAACAIoggAAAAQBFEEAAAAKAIIggAAABQBBEEAAAAKIIIAgAAABRBBAEAAACKIIIAAAAARRBBAAAAgCKIIAAAAEARRBAAAACgCCIIAAAAUAQRBAAAACjCiFoP2LhxY9auXZtKpZK5c+dm3rx5tZ4EAAAAHINqeiZIpVLJvffemy9+8YtZuXJlHn/88bz66qu1nAQAAAAco2oaQV588cWcdNJJ+a3f+q2MGDEiF1xwQbq6umo5CQAAADhG1fX39/fX6sX/7d/+LRs3bsx1112XJHn00UezdevWXHPNNQd9X2dnZzo7O5Mk7e3tVd8JAAAADH81PRPkUP2lrq7uPfe1tbWlvb1dABkGli1bVusJMGw5fuDIOHbgyDl+4Mg5foanmkaQiRMnZseOHQdu79ixI01NTTVcBAAAAByrahpBPvjBD+b111/Ptm3b0tfXlyeeeCLnnXdeLScBAAAAx6iafkRuQ0NDFi9enK9+9aupVCq55JJLcsopp9RyEu9TW1tbrSfAsOX4gSPj2IEj5/iBI+f4GZ5q+saoAAAAANVS08thAAAAAKpFBAEAAACKUNP3BOHYcdddd+Wpp55KY2NjVqxYUes5MGx0d3dn9erVefPNN1NXV5e2trZcdtlltZ4Fw0Jvb2/uuOOO9PX1Zf/+/Zk9e3auvPLKWs+CYaNSqWTZsmWZMGGCj/qEw/D5z38+xx9/fOrr69PQ0JD29vZaT+IwiCAMiosvvji///u/n9WrV9d6CgwrDQ0NWbhwYaZMmZK9e/dm2bJl+chHPpLW1tZaT4Oj3nHHHZc77rgjxx9/fPr6+nL77bfn7LPPzvTp02s9DYaFH/zgB5k8eXL27t1b6ykw7Nxxxx0ZN25crWdwBFwOw6A4/fTTM3bs2FrPgGGnqakpU6ZMSZKMHj06kydPzs6dO2u8CoaHurq6HH/88UmS/fv3Z//+/amrq6vxKhgeduzYkaeeeipz586t9RSAqnImCMBRYtu2bXnppZcyderUWk+BYaNSqeTWW2/N//zP/+TjH/94pk2bVutJMCysW7cuCxYscBYIHKGvfvWrSZKPfexjPip3mBFBAI4C+/bty4oVK7Jo0aKMGTOm1nNg2Kivr8/Xv/71vPXWW1m+fHlefvnlnHrqqbWeBUe1H//4x2lsbMyUKVOyadOmWs+BYefLX/5yJkyYkF27duUrX/lKWlpacvrpp9d6FgMkggDUWF9fX1asWJGLLroos2bNqvUcGJZOOOGEnH766dm4caMIAr/Bli1b8uSTT2bDhg3p7e3N3r17s2rVqixdurTW02BYmDBhQpKksbExM2fOzIsvviiCDCMiCEAN9ff355577snkyZPzyU9+stZzYFjZvXt3GhoacsIJJ6S3tzfPPvtsPv3pT9d6Fhz15s+fn/nz5ydJNm3alIcfflgAgQHat29f+vv7M3r06Ozbty/PPPNM/uiP/qjWszgMIgiD4pvf/Gaef/75/OxnP8t1112XK6+8MpdeemmtZ8FRb8uWLXn00Udz6qmn5pZbbkmSXHXVVTnnnHNqvAyOfj09PVm9enUqlUr6+/tz/vnn59xzz631LACOYbt27cry5cuT/OJNuS+88MKcffbZtR3FYanr7+/vr/UIAAAAgKHmI3IBAACAIoggAAAAQBFEEAAAAKAIIggAAABQBBEEAAAAKIIIAgAUb+HChXnjjTdqPQMAGGI+IhcAeF/+8z//M9/61rfyyiuvpL6+Pq2trfnMZz6TqVOn5kc/+lH+9V//NV/+8pcH7fU2bdqUL33pSxk5cmTq6urS1NSUefPm5ZJLLhnQ4//sz/4sF110UebOnTtomwCA4WFErQcAAMPXz3/+87S3t+faa6/NBRdckL6+vmzevDnHHXfcoDz//v3709DQ8J77m5qacs8996S/vz8bNmzIX/7lX+ZDH/pQWlpaBuV1AYBjkwgCAByx119/PUly4YUXJklGjhyZs846K0ny6quv5m/+5m/S19eXhQsXpqGhIevWrctTTz2VBx54IG+88UbGjBmTSy65JFdeeWWSZNu2bbnhhhty3XXX5cEHH8yJJ56YP//zP/+Vr19XV5dzzjknY8eOzU9+8pO0tLRkz549ufPOO7N169ZUKpV86EMfyuc+97lMnDgxf/d3f5fNmzdn69atWbduXS6++OJcc801ufLKK7Nq1aqcdNJJWb16dUaNGpXt27dn8+bNaW1tzdKlS3PSSSclSZ5++umsWbMmb775Zi666KK88sor+b3f+z1nlgDAMOA9QQCAI3byySenvr4+d955ZzZs2JA9e/Yc+Fpra2s+97nPZfr06bn//vuzbt26JMmoUaNyww03ZO3atVm2bFn+5V/+Jf/xH/9x0PM+//zzWblyZW677bZf+/qVSiVPPvlkfvaznx2IFP39/bn44otz11135a677srIkSNz7733JkmuuuqqfPjDH87ixYtz//3355prrjnk8z7++OO54oorsnbt2px00kl54IEHkiS7d+/ON77xjcyfPz9r1qxJS0tLXnjhhSP6bwcAVJ8IAgAcsTFjxuRLX/pS6urq8td//de59tpr87WvfS1vvvnmr3zMGWeckVNPPTX19fX57d/+7Xz0ox/N888/f9D3XHHFFTn++OMzcuTIQz5HT09PFi1alAULFuTrX/96rr766vzu7/5ukuQDH/hAZs+enVGjRmX06NG5/PLLs3nz5sP6fc2aNStTp05NQ0NDLrzwwvz3f/93kmTDhg1pbW3NrFmz0tDQkD/4gz/I+PHjD+u5AYDacTkMAPC+tLa25vOf/3yS5Kc//Wn+6q/+KuvWrcuf/MmfHPL7t27dmr/927/Nyy+/nL6+vvT19WX27NkHfc/EiRN/7Wu++54g77zzTr797W/nueeeyyc+8Ykkydtvv5377rsvGzduzFtvvZUk2bt3byqVSurrB/b/f345bIwaNSr79u1L8ov48svb6urqMmHChAE9JwBQe84EAQAGzeTJk3PxxRfnlVde+ZXfs2rVqpx77rm5++67c9999+VjH/tY/veH1dXV1Q3o9Y477rgsWLAgL7/88oFLah5++OG89tpr+Yu/+Ivcd999B95T5N3XGOhzH8r48eOzc+fOA7f7+/sPug0AHN1EEADgiP30pz/Nww8/nB07diRJuru78/jjj2fatGlJ/n806OvrO/CYvXv3ZuzYsRk5cmRefPHFPPbYY+9rw4gRI/KpT30qDz30UJJk3759GTlyZMaMGZM9e/bkwQcfPOj7Gxsb88YbbxzRa51zzjkHgsv+/fvzwx/+8Nde+gMAHF1EEADgiI0ePTpbt27NF7/4xSxcuDC33XZbTjnllFx99dVJkjPPPPPAG6S++yak1157bb7zne/k6quvzj/8wz/k/PPPf987LrnkknR3d+fJJ5/MZZddlt7e3lxzzTW57bbbcvbZZx/0vZdddln+/d//PZ/97GezZs2aw3qdcePG5eabb863v/3tLF68OK+++mqmTJkyaB8JDAAMrbr+/33+KQAAA1KpVLJkyZLceOONOfPMM2s9BwD4DZwJAgBwGN59w9V33nkn3/3ud9Pf35/p06fXehYAMAA+HQYA4DC88MILWbVqVfr6+tLa2ppbbrnlV36ULwBwdHE5DAAAAFAEl8MAAAAARRBBAAAAgCKIIAAAAEARRBAAAACgCCIIAAAAUIT/B/NUL/cNJfJlAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sb.factorplot('Star Rating',data=dt,kind=\"count\", aspect=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7dcf5d8c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\vicen\\anaconda3\\lib\\site-packages\\seaborn\\categorical.py:3714: UserWarning: The `factorplot` function has been renamed to `catplot`. The original name will be removed in a future release. Please update your code. Note that the default `kind` in `factorplot` (`'point'`) has changed `'strip'` in `catplot`.\n",
      "  warnings.warn(msg)\n",
      "C:\\Users\\vicen\\anaconda3\\lib\\site-packages\\seaborn\\_decorators.py:36: FutureWarning: Pass the following variable as a keyword arg: x. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x15f17614e80>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sb.factorplot('wordcount',data=dt,kind=\"count\", aspect=3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d18ca8cf",
   "metadata": {},
   "source": [
    "Creamos nuestro conjuntos de entrenamiento y test para una validación honesta de nuestro modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c25ee234",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = dt[['wordcount','sentimentValue']].values\n",
    "y = dt['Star Rating'].values\n",
    " \n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0, test_size=0.3)\n",
    "scaler = MinMaxScaler()\n",
    "X_train = scaler.fit_transform(X_train)\n",
    "X_test = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b1a0a42",
   "metadata": {},
   "source": [
    "Usamos la librería scikit-learn que tiene muchos algoritmos de machine learning para usar el algoritmos KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3e567421",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of K-NN classifier on training set: 0.89\n",
      "Accuracy of K-NN classifier on test set: 0.87\n"
     ]
    }
   ],
   "source": [
    "n_neighbors = 7\n",
    " \n",
    "knn = KNeighborsClassifier(n_neighbors)\n",
    "knn.fit(X_train, y_train)\n",
    "print('Accuracy of K-NN classifier on training set: {:.2f}'\n",
    "     .format(knn.score(X_train, y_train)))\n",
    "print('Accuracy of K-NN classifier on test set: {:.2f}'\n",
    "     .format(knn.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ff01c04",
   "metadata": {},
   "source": [
    "Vemos que la precisión que nos da es de 87% en el set de entrenamiento y del 86% para el de test.\n",
    "\n",
    "NOTA: como verán utilizamos la clase KNeighborsClassifier de SciKit Learn puesto que nuestras etiquetas son valores discretos (estrellas del 1 al 5). Pero deben saber que también existe la clase KneighborsRegressor para etiquetas con valores continuos."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01ffb2a2",
   "metadata": {},
   "source": [
    "### Métricas del modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "13c54684",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[10  0  1  0  0]\n",
      " [ 0  2  0  0  0]\n",
      " [ 0  1 22  0  1]\n",
      " [ 0  0  2  8  1]\n",
      " [ 0  0  4  0 26]]\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       1.00      0.91      0.95        11\n",
      "           2       0.67      1.00      0.80         2\n",
      "           3       0.76      0.92      0.83        24\n",
      "           4       1.00      0.73      0.84        11\n",
      "           5       0.93      0.87      0.90        30\n",
      "\n",
      "    accuracy                           0.87        78\n",
      "   macro avg       0.87      0.88      0.86        78\n",
      "weighted avg       0.89      0.87      0.87        78\n",
      "\n"
     ]
    }
   ],
   "source": [
    "pred = knn.predict(X_test)\n",
    "print(confusion_matrix(y_test, pred))\n",
    "print(classification_report(y_test, pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d0a160ac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8717948717948718"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(y_test, pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35a5ab68",
   "metadata": {},
   "source": [
    "Como se ve la puntuación F1 es del 87%, bastante buena. NOTA: recuerden que este es sólo un ejercicio para aprender y tenemos MUY pocos registros totales y en nuestro conjunto de test. Por ejemplo de 2 estrellas sólo tiene 1 valoración y esto es evidentemente insuficiente."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2e21d839",
   "metadata": {},
   "source": [
    "Ahora realizaremos la grafica con la clasificación obtenida, la que nos ayuda a ver fácilmente en donde caerán las predicciones. NOTA: al ser 2 features, podemos hacer la gráfica 2D y si fueran 3 podría ser en 3D. Pero para usos reales, podríamos tener  más de 3 dimensiones y no importaría  poder visualizarlo sino el resultado del algoritmo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "5603cd2a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "h = .02  # step size in the mesh\n",
    " \n",
    "# Create color maps\n",
    "cmap_light = ListedColormap(['#FFAAAA', '#ffcc99', '#ffffb3','#b3ffff','#c2f0c2'])\n",
    "cmap_bold = ListedColormap(['#FF0000', '#ff9933','#FFFF00','#00ffff','#00FF00'])\n",
    " \n",
    "# we create an instance of Neighbours Classifier and fit the data.\n",
    "clf = KNeighborsClassifier(n_neighbors, weights='distance')\n",
    "clf.fit(X, y)\n",
    " \n",
    "# Plot the decision boundary. For that, we will assign a color to each\n",
    "# point in the mesh [x_min, x_max]x[y_min, y_max].\n",
    "x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1\n",
    "y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, h),\n",
    "                         np.arange(y_min, y_max, h))\n",
    "Z = clf.predict(np.c_[xx.ravel(), yy.ravel()])\n",
    " \n",
    "# Put the result into a color plot\n",
    "Z = Z.reshape(xx.shape)\n",
    "plt.figure()\n",
    "plt.pcolormesh(xx, yy, Z, cmap=cmap_light)\n",
    " \n",
    "# Plot also the training points\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, cmap=cmap_bold,\n",
    "                edgecolor='k', s=20)\n",
    "plt.xlim(xx.min(), xx.max())\n",
    "plt.ylim(yy.min(), yy.max())\n",
    "    \n",
    "patch0 = mpatches.Patch(color='#FF0000', label='1')\n",
    "patch1 = mpatches.Patch(color='#ff9933', label='2')\n",
    "patch2 = mpatches.Patch(color='#FFFF00', label='3')\n",
    "patch3 = mpatches.Patch(color='#00ffff', label='4')\n",
    "patch4 = mpatches.Patch(color='#00FF00', label='5')\n",
    "plt.legend(handles=[patch0, patch1, patch2, patch3,patch4])\n",
    " \n",
    "    \n",
    "plt.title(\"5-Class classification (k = %i, weights = '%s')\"\n",
    "              % (n_neighbors, 'distance'))\n",
    " \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f4d53a0",
   "metadata": {},
   "source": [
    "Es decir que “a ojo” una review de 20 palabras y Sentimiento 1, nos daría una valoración de 4 (zona celeste).\n",
    "\n",
    "Con estas zonas podemos intuir ciertas características de los usuarios que usan y valoran la app:\n",
    "\n",
    "- Los usuarios que ponen 1 estrella tienen sentimiento negativo y hasta 25 palabras.\n",
    "- Los usuarios que ponen 2 estrellas dan muchas explicaciones (hasta 100 palabras) y su sentimiento puede variar entre negativo y algo positivo.\n",
    "- Los usuarios que ponen 3 estrellas son bastante neutrales en sentimientos, puesto que están en torno al cero y hasta unas 25 palabras.\n",
    "- Los usuarios que dan 5 estrellas son bastante positivos (de 0,5 en adelante, aproximadamente) y ponen pocas palabras (hasta 10)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca02efbe",
   "metadata": {},
   "source": [
    "### Ejercicio\n",
    "\n",
    "Ahora que sabéis implementar el KNN, me gustaría que hicierais un pequeño código aprovechando lo que habéis usado arriba para averiguar cual es el valor de K que hace que mejore el score f1. Probad los valores entre 1 y 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "67f990ad",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "a311bfce",
   "metadata": {},
   "source": [
    "### Ejercicio\n",
    "\n",
    "Usando el código del primer ejercicio implementa el KNN de K=1. Qué scores f1 obtenemos?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ee80042a",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "58c40fb0",
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}