ML_model_collage/model_collage.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se cargan las librerías\n",
    "library(\"dplyr\")\n",
    "library(\"readr\")\n",
    "library(\"ggplot2\")\n",
    "# Paquete que contiene la función sample.split\n",
    "library(\"caTools\")\n",
    "# Paquete que contiene la función para el modelo de análisis discriminante\n",
    "library(\"MASS\")\n",
    "# Paquete que contiene la función para el modelo de k-vecinos\n",
    "library(class)\n",
    "# Paquete que contiene la función para las gráficas ROC\n",
    "library(\"ROCR\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Lee la información de los archivos\n",
    "usuarios <- read.table(\"./movies_data/u.user\", header = TRUE, sep = \"|\")\n",
    "peliculas <- read.table(\"./movies_data/u.item\", header = TRUE, sep = \"|\")\n",
    "calificaciones <- read.table(\"./movies_data/udata.data\", header = TRUE, sep = \"\\t\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se crea una tabla con la información de los usuarios y sus calificaciones\n",
    "info <- left_join(usuarios, calificaciones, by = \"userid\", multiple = \"all\")\n",
    "info <- left_join(info, peliculas, by = \"itemid\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se crea una función para calcular el porcentaje de aciertos\n",
    "precision <- function(x) {\n",
    "    sum(diag(x) / (sum(rowSums(x)))) * 100\n",
    "}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"dataframe\">\n",
       "<caption>A data.frame: 6 × 31</caption>\n",
       "<thead>\n",
       "\t<tr><th></th><th scope=col>userid</th><th scope=col>age</th><th scope=col>gender</th><th scope=col>occupation</th><th scope=col>zcode</th><th scope=col>itemid</th><th scope=col>rate</th><th scope=col>timestamp</th><th scope=col>title</th><th scope=col>rdate</th><th scope=col>⋯</th><th scope=col>Fantasy</th><th scope=col>FilmNoir</th><th scope=col>Horror</th><th scope=col>Musical</th><th scope=col>Mystery</th><th scope=col>Romance</th><th scope=col>SciFi</th><th scope=col>Thriller</th><th scope=col>War</th><th scope=col>Western</th></tr>\n",
       "\t<tr><th></th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>⋯</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><th scope=row>1</th><td>1</td><td>24</td><td>M</td><td>technician</td><td>85711</td><td> 61</td><td>4</td><td>878542420</td><td>Three Colors: White (1994)</td><td>01-Jan-1994</td><td>⋯</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td></tr>\n",
       "\t<tr><th scope=row>2</th><td>1</td><td>24</td><td>M</td><td>technician</td><td>85711</td><td>189</td><td>3</td><td>888732928</td><td>Grand Day Out, A (1992)   </td><td>01-Jan-1992</td><td>⋯</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td></tr>\n",
       "\t<tr><th scope=row>3</th><td>1</td><td>24</td><td>M</td><td>technician</td><td>85711</td><td> 33</td><td>4</td><td>878542699</td><td>Desperado (1995)          </td><td>01-Jan-1995</td><td>⋯</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>1</td><td>0</td><td>1</td><td>0</td><td>0</td></tr>\n",
       "\t<tr><th scope=row>4</th><td>1</td><td>24</td><td>M</td><td>technician</td><td>85711</td><td>160</td><td>4</td><td>875072547</td><td>Glengarry Glen Ross (1992)</td><td>01-Jan-1992</td><td>⋯</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td></tr>\n",
       "\t<tr><th scope=row>5</th><td>1</td><td>24</td><td>M</td><td>technician</td><td>85711</td><td> 20</td><td>4</td><td>887431883</td><td>Angels and Insects (1995) </td><td>01-Jan-1995</td><td>⋯</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>1</td><td>0</td><td>0</td><td>0</td><td>0</td></tr>\n",
       "\t<tr><th scope=row>6</th><td>1</td><td>24</td><td>M</td><td>technician</td><td>85711</td><td>202</td><td>5</td><td>875072442</td><td>Groundhog Day (1993)      </td><td>01-Jan-1993</td><td>⋯</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>1</td><td>0</td><td>0</td><td>0</td><td>0</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "A data.frame: 6 × 31\n",
       "\\begin{tabular}{r|lllllllllllllllllllll}\n",
       "  & userid & age & gender & occupation & zcode & itemid & rate & timestamp & title & rdate & ⋯ & Fantasy & FilmNoir & Horror & Musical & Mystery & Romance & SciFi & Thriller & War & Western\\\\\n",
       "  & <int> & <int> & <chr> & <chr> & <chr> & <int> & <int> & <dbl> & <chr> & <chr> & ⋯ & <int> & <int> & <int> & <int> & <int> & <int> & <int> & <int> & <int> & <int>\\\\\n",
       "\\hline\n",
       "\t1 & 1 & 24 & M & technician & 85711 &  61 & 4 & 878542420 & Three Colors: White (1994) & 01-Jan-1994 & ⋯ & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0\\\\\n",
       "\t2 & 1 & 24 & M & technician & 85711 & 189 & 3 & 888732928 & Grand Day Out, A (1992)    & 01-Jan-1992 & ⋯ & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0\\\\\n",
       "\t3 & 1 & 24 & M & technician & 85711 &  33 & 4 & 878542699 & Desperado (1995)           & 01-Jan-1995 & ⋯ & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 1 & 0 & 0\\\\\n",
       "\t4 & 1 & 24 & M & technician & 85711 & 160 & 4 & 875072547 & Glengarry Glen Ross (1992) & 01-Jan-1992 & ⋯ & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0\\\\\n",
       "\t5 & 1 & 24 & M & technician & 85711 &  20 & 4 & 887431883 & Angels and Insects (1995)  & 01-Jan-1995 & ⋯ & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0\\\\\n",
       "\t6 & 1 & 24 & M & technician & 85711 & 202 & 5 & 875072442 & Groundhog Day (1993)       & 01-Jan-1993 & ⋯ & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "A data.frame: 6 × 31\n",
       "\n",
       "| <!--/--> | userid &lt;int&gt; | age &lt;int&gt; | gender &lt;chr&gt; | occupation &lt;chr&gt; | zcode &lt;chr&gt; | itemid &lt;int&gt; | rate &lt;int&gt; | timestamp &lt;dbl&gt; | title &lt;chr&gt; | rdate &lt;chr&gt; | ⋯ ⋯ | Fantasy &lt;int&gt; | FilmNoir &lt;int&gt; | Horror &lt;int&gt; | Musical &lt;int&gt; | Mystery &lt;int&gt; | Romance &lt;int&gt; | SciFi &lt;int&gt; | Thriller &lt;int&gt; | War &lt;int&gt; | Western &lt;int&gt; |\n",
       "|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n",
       "| 1 | 1 | 24 | M | technician | 85711 |  61 | 4 | 878542420 | Three Colors: White (1994) | 01-Jan-1994 | ⋯ | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |\n",
       "| 2 | 1 | 24 | M | technician | 85711 | 189 | 3 | 888732928 | Grand Day Out, A (1992)    | 01-Jan-1992 | ⋯ | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |\n",
       "| 3 | 1 | 24 | M | technician | 85711 |  33 | 4 | 878542699 | Desperado (1995)           | 01-Jan-1995 | ⋯ | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 1 | 0 | 0 |\n",
       "| 4 | 1 | 24 | M | technician | 85711 | 160 | 4 | 875072547 | Glengarry Glen Ross (1992) | 01-Jan-1992 | ⋯ | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |\n",
       "| 5 | 1 | 24 | M | technician | 85711 |  20 | 4 | 887431883 | Angels and Insects (1995)  | 01-Jan-1995 | ⋯ | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 |\n",
       "| 6 | 1 | 24 | M | technician | 85711 | 202 | 5 | 875072442 | Groundhog Day (1993)       | 01-Jan-1993 | ⋯ | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 |\n",
       "\n"
      ],
      "text/plain": [
       "  userid age gender occupation zcode itemid rate timestamp\n",
       "1 1      24  M      technician 85711  61    4    878542420\n",
       "2 1      24  M      technician 85711 189    3    888732928\n",
       "3 1      24  M      technician 85711  33    4    878542699\n",
       "4 1      24  M      technician 85711 160    4    875072547\n",
       "5 1      24  M      technician 85711  20    4    887431883\n",
       "6 1      24  M      technician 85711 202    5    875072442\n",
       "  title                      rdate       ⋯ Fantasy FilmNoir Horror Musical\n",
       "1 Three Colors: White (1994) 01-Jan-1994 ⋯ 0       0        0      0      \n",
       "2 Grand Day Out, A (1992)    01-Jan-1992 ⋯ 0       0        0      0      \n",
       "3 Desperado (1995)           01-Jan-1995 ⋯ 0       0        0      0      \n",
       "4 Glengarry Glen Ross (1992) 01-Jan-1992 ⋯ 0       0        0      0      \n",
       "5 Angels and Insects (1995)  01-Jan-1995 ⋯ 0       0        0      0      \n",
       "6 Groundhog Day (1993)       01-Jan-1993 ⋯ 0       0        0      0      \n",
       "  Mystery Romance SciFi Thriller War Western\n",
       "1 0       0       0     0        0   0      \n",
       "2 0       0       0     0        0   0      \n",
       "3 0       1       0     1        0   0      \n",
       "4 0       0       0     0        0   0      \n",
       "5 0       1       0     0        0   0      \n",
       "6 0       1       0     0        0   0      "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "head(info)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>\n",
       ".list-inline {list-style: none; margin:0; padding: 0}\n",
       ".list-inline>li {display: inline-block}\n",
       ".list-inline>li:not(:last-child)::after {content: \"\\00b7\"; padding: 0 .5ex}\n",
       "</style>\n",
       "<ol class=list-inline><li>'userid'</li><li>'age'</li><li>'gender'</li><li>'occupation'</li><li>'zcode'</li><li>'itemid'</li><li>'rate'</li><li>'timestamp'</li><li>'title'</li><li>'rdate'</li><li>'vrdate'</li><li>'IMDbURL'</li><li>'unknown'</li><li>'Action'</li><li>'Adventure'</li><li>'Animation'</li><li>'Children'</li><li>'Comedy'</li><li>'Crime'</li><li>'Documentary'</li><li>'Drama'</li><li>'Fantasy'</li><li>'FilmNoir'</li><li>'Horror'</li><li>'Musical'</li><li>'Mystery'</li><li>'Romance'</li><li>'SciFi'</li><li>'Thriller'</li><li>'War'</li><li>'Western'</li></ol>\n"
      ],
      "text/latex": [
       "\\begin{enumerate*}\n",
       "\\item 'userid'\n",
       "\\item 'age'\n",
       "\\item 'gender'\n",
       "\\item 'occupation'\n",
       "\\item 'zcode'\n",
       "\\item 'itemid'\n",
       "\\item 'rate'\n",
       "\\item 'timestamp'\n",
       "\\item 'title'\n",
       "\\item 'rdate'\n",
       "\\item 'vrdate'\n",
       "\\item 'IMDbURL'\n",
       "\\item 'unknown'\n",
       "\\item 'Action'\n",
       "\\item 'Adventure'\n",
       "\\item 'Animation'\n",
       "\\item 'Children'\n",
       "\\item 'Comedy'\n",
       "\\item 'Crime'\n",
       "\\item 'Documentary'\n",
       "\\item 'Drama'\n",
       "\\item 'Fantasy'\n",
       "\\item 'FilmNoir'\n",
       "\\item 'Horror'\n",
       "\\item 'Musical'\n",
       "\\item 'Mystery'\n",
       "\\item 'Romance'\n",
       "\\item 'SciFi'\n",
       "\\item 'Thriller'\n",
       "\\item 'War'\n",
       "\\item 'Western'\n",
       "\\end{enumerate*}\n"
      ],
      "text/markdown": [
       "1. 'userid'\n",
       "2. 'age'\n",
       "3. 'gender'\n",
       "4. 'occupation'\n",
       "5. 'zcode'\n",
       "6. 'itemid'\n",
       "7. 'rate'\n",
       "8. 'timestamp'\n",
       "9. 'title'\n",
       "10. 'rdate'\n",
       "11. 'vrdate'\n",
       "12. 'IMDbURL'\n",
       "13. 'unknown'\n",
       "14. 'Action'\n",
       "15. 'Adventure'\n",
       "16. 'Animation'\n",
       "17. 'Children'\n",
       "18. 'Comedy'\n",
       "19. 'Crime'\n",
       "20. 'Documentary'\n",
       "21. 'Drama'\n",
       "22. 'Fantasy'\n",
       "23. 'FilmNoir'\n",
       "24. 'Horror'\n",
       "25. 'Musical'\n",
       "26. 'Mystery'\n",
       "27. 'Romance'\n",
       "28. 'SciFi'\n",
       "29. 'Thriller'\n",
       "30. 'War'\n",
       "31. 'Western'\n",
       "\n",
       "\n"
      ],
      "text/plain": [
       " [1] \"userid\"      \"age\"         \"gender\"      \"occupation\"  \"zcode\"      \n",
       " [6] \"itemid\"      \"rate\"        \"timestamp\"   \"title\"       \"rdate\"      \n",
       "[11] \"vrdate\"      \"IMDbURL\"     \"unknown\"     \"Action\"      \"Adventure\"  \n",
       "[16] \"Animation\"   \"Children\"    \"Comedy\"      \"Crime\"       \"Documentary\"\n",
       "[21] \"Drama\"       \"Fantasy\"     \"FilmNoir\"    \"Horror\"      \"Musical\"    \n",
       "[26] \"Mystery\"     \"Romance\"     \"SciFi\"       \"Thriller\"    \"War\"        \n",
       "[31] \"Western\"    "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colnames(info)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Planteamiento para modelo de regresión logística\n",
    "# ¿Cómo puedo saber si una calificación fue hecha por un hombre o una mujer?\n",
    "\n",
    "# Se agrega una columna con el género del usuario en forma de bit, 1 para mujer y 0 para hombre,\n",
    "# esto con el fin de poder utilizarlo en el modelo de regresión logística\n",
    "info$gender_as_bit <- ifelse(info$gender == \"F\", 1, 0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se separan los datos en entrenamiento y prueba\n",
    "separador_RL <- sample.split(info, SplitRatio = 0.8)\n",
    "info_entrenamiento_RL <- info[separador_RL, ]\n",
    "info_prueba_RL <- info[-separador_RL, ]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se crea el modelo de regresión logística\n",
    "\n",
    "modelo_RL <- glm(gender_as_bit ~ itemid + rate + unknown + Action + Adventure + Animation + Children + Comedy + Crime + Documentary + Drama + Fantasy + FilmNoir + Horror + Musical + Mystery + Romance + SciFi + Thriller + War + Western,\n",
    "    data = info_entrenamiento_RL,\n",
    "    family = binomial\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se obtienen 3 registros de prueba, sin la columna de género\n",
    "reg_prueba_RL <- info_prueba_RL[1:3, -info_prueba_RL$gender_as_bit]\n",
    "# Se obtienen las probabilidades de que el usuario sea hombre o mujer (la predicción)\n",
    "prediccion_RL <- predict(modelo_RL, reg_prueba_RL, type = \"response\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"dataframe\">\n",
       "<caption>A data.frame: 3 × 31</caption>\n",
       "<thead>\n",
       "\t<tr><th></th><th scope=col>age</th><th scope=col>gender</th><th scope=col>occupation</th><th scope=col>zcode</th><th scope=col>itemid</th><th scope=col>rate</th><th scope=col>timestamp</th><th scope=col>title</th><th scope=col>rdate</th><th scope=col>vrdate</th><th scope=col>⋯</th><th scope=col>FilmNoir</th><th scope=col>Horror</th><th scope=col>Musical</th><th scope=col>Mystery</th><th scope=col>Romance</th><th scope=col>SciFi</th><th scope=col>Thriller</th><th scope=col>War</th><th scope=col>Western</th><th scope=col>gender_as_bit</th></tr>\n",
       "\t<tr><th></th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;lgl&gt;</th><th scope=col>⋯</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;dbl&gt;</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><th scope=row>2</th><td>24</td><td>M</td><td>technician</td><td>85711</td><td>189</td><td>3</td><td>888732928</td><td>Grand Day Out, A (1992)   </td><td>01-Jan-1992</td><td>NA</td><td>⋯</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td></tr>\n",
       "\t<tr><th scope=row>3</th><td>24</td><td>M</td><td>technician</td><td>85711</td><td> 33</td><td>4</td><td>878542699</td><td>Desperado (1995)          </td><td>01-Jan-1995</td><td>NA</td><td>⋯</td><td>0</td><td>0</td><td>0</td><td>0</td><td>1</td><td>0</td><td>1</td><td>0</td><td>0</td><td>0</td></tr>\n",
       "\t<tr><th scope=row>4</th><td>24</td><td>M</td><td>technician</td><td>85711</td><td>160</td><td>4</td><td>875072547</td><td>Glengarry Glen Ross (1992)</td><td>01-Jan-1992</td><td>NA</td><td>⋯</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "A data.frame: 3 × 31\n",
       "\\begin{tabular}{r|lllllllllllllllllllll}\n",
       "  & age & gender & occupation & zcode & itemid & rate & timestamp & title & rdate & vrdate & ⋯ & FilmNoir & Horror & Musical & Mystery & Romance & SciFi & Thriller & War & Western & gender\\_as\\_bit\\\\\n",
       "  & <int> & <chr> & <chr> & <chr> & <int> & <int> & <dbl> & <chr> & <chr> & <lgl> & ⋯ & <int> & <int> & <int> & <int> & <int> & <int> & <int> & <int> & <int> & <dbl>\\\\\n",
       "\\hline\n",
       "\t2 & 24 & M & technician & 85711 & 189 & 3 & 888732928 & Grand Day Out, A (1992)    & 01-Jan-1992 & NA & ⋯ & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0\\\\\n",
       "\t3 & 24 & M & technician & 85711 &  33 & 4 & 878542699 & Desperado (1995)           & 01-Jan-1995 & NA & ⋯ & 0 & 0 & 0 & 0 & 1 & 0 & 1 & 0 & 0 & 0\\\\\n",
       "\t4 & 24 & M & technician & 85711 & 160 & 4 & 875072547 & Glengarry Glen Ross (1992) & 01-Jan-1992 & NA & ⋯ & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "A data.frame: 3 × 31\n",
       "\n",
       "| <!--/--> | age &lt;int&gt; | gender &lt;chr&gt; | occupation &lt;chr&gt; | zcode &lt;chr&gt; | itemid &lt;int&gt; | rate &lt;int&gt; | timestamp &lt;dbl&gt; | title &lt;chr&gt; | rdate &lt;chr&gt; | vrdate &lt;lgl&gt; | ⋯ ⋯ | FilmNoir &lt;int&gt; | Horror &lt;int&gt; | Musical &lt;int&gt; | Mystery &lt;int&gt; | Romance &lt;int&gt; | SciFi &lt;int&gt; | Thriller &lt;int&gt; | War &lt;int&gt; | Western &lt;int&gt; | gender_as_bit &lt;dbl&gt; |\n",
       "|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n",
       "| 2 | 24 | M | technician | 85711 | 189 | 3 | 888732928 | Grand Day Out, A (1992)    | 01-Jan-1992 | NA | ⋯ | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |\n",
       "| 3 | 24 | M | technician | 85711 |  33 | 4 | 878542699 | Desperado (1995)           | 01-Jan-1995 | NA | ⋯ | 0 | 0 | 0 | 0 | 1 | 0 | 1 | 0 | 0 | 0 |\n",
       "| 4 | 24 | M | technician | 85711 | 160 | 4 | 875072547 | Glengarry Glen Ross (1992) | 01-Jan-1992 | NA | ⋯ | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |\n",
       "\n"
      ],
      "text/plain": [
       "  age gender occupation zcode itemid rate timestamp title                     \n",
       "2 24  M      technician 85711 189    3    888732928 Grand Day Out, A (1992)   \n",
       "3 24  M      technician 85711  33    4    878542699 Desperado (1995)          \n",
       "4 24  M      technician 85711 160    4    875072547 Glengarry Glen Ross (1992)\n",
       "  rdate       vrdate ⋯ FilmNoir Horror Musical Mystery Romance SciFi Thriller\n",
       "2 01-Jan-1992 NA     ⋯ 0        0      0       0       0       0     0       \n",
       "3 01-Jan-1995 NA     ⋯ 0        0      0       0       1       0     1       \n",
       "4 01-Jan-1992 NA     ⋯ 0        0      0       0       0       0     0       \n",
       "  War Western gender_as_bit\n",
       "2 0   0       0            \n",
       "3 0   0       0            \n",
       "4 0   0       0            "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<style>\n",
       ".dl-inline {width: auto; margin:0; padding: 0}\n",
       ".dl-inline>dt, .dl-inline>dd {float: none; width: auto; display: inline-block}\n",
       ".dl-inline>dt::after {content: \":\\0020\"; padding-right: .5ex}\n",
       ".dl-inline>dt:not(:first-of-type) {padding-left: .5ex}\n",
       "</style><dl class=dl-inline><dt>2</dt><dd>'M'</dd><dt>3</dt><dd>'M'</dd><dt>4</dt><dd>'M'</dd></dl>\n"
      ],
      "text/latex": [
       "\\begin{description*}\n",
       "\\item[2] 'M'\n",
       "\\item[3] 'M'\n",
       "\\item[4] 'M'\n",
       "\\end{description*}\n"
      ],
      "text/markdown": [
       "2\n",
       ":   'M'3\n",
       ":   'M'4\n",
       ":   'M'\n",
       "\n"
      ],
      "text/plain": [
       "  2   3   4 \n",
       "\"M\" \"M\" \"M\" "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Se obtiene la predicción final\n",
    "prediccion_RL <- ifelse(prediccion_RL > 0.5, \"F\", \"M\")\n",
    "reg_prueba_RL\n",
    "prediccion_RL\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "Plot with title \"Curva ROC para el modelo de regresión logística\""
      ]
     },
     "metadata": {
      "image/png": {
       "height": 420,
       "width": 420
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Se hace una predicción para todos los registros de prueba\n",
    "prediccion_RL <- predict(modelo_RL, info_prueba_RL, type = \"response\")\n",
    "# Se usa la gráfica ROC para evaluar el modelo\n",
    "prediccion_RL <- prediction(prediccion_RL, info_prueba_RL$gender_as_bit)\n",
    "perf_RL <- performance(prediccion_RL, measure = \"tpr\", x.measure = \"fpr\")\n",
    "plot(perf_RL, colorize = TRUE, main = \"Curva ROC para el modelo de regresión logística\")\n",
    "abline(0, 1, lty = 2)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A través de esta gráfica, se puede apreciar que el rendimiento de modelo es pobre, pues al probar con múltiples umbrales (thresholds) se nota como el modelo incrementa tu tasa de verdaderos positivos (cuando predice que la calificación fue dada por un usuario mujer y está en lo correcto), pero también su tasa de falsos positivos (se equivoca al decir que una calificación fue hecha por una mujer, cuando no fue así)\n",
    "\n",
    "De esto se puede terminar que, a partir de la complejidad de los datos, determinar si la calificación fue hecha por un hombre o una mujer no es una tarea factible."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "73"
      ],
      "text/latex": [
       "73"
      ],
      "text/markdown": [
       "73"
      ],
      "text/plain": [
       "[1] 73"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "7"
      ],
      "text/latex": [
       "7"
      ],
      "text/markdown": [
       "7"
      ],
      "text/plain": [
       "[1] 7"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Planteamiento para modelo de análisis discriminante\n",
    "# Se desea saber el grupo etario de un usuario a partir de su ocupación, calificaciones,\n",
    "# género de las películas (se seleccionan 5) y sexo\n",
    "\n",
    "# Revisamos la edad máxima y mínima de los usuarios\n",
    "max(info$age)\n",
    "min(info$age)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se crea una columna con el grupo etario del usuario, \"ñ\" para niños, \"a\" para adolescentes, \"A\" para adultos y \"AM\" para adultos mayores\n",
    "info$age_group <- ifelse(info$age < 10, \"ñ\",\n",
    "    ifelse(info$age < 18, \"a\",\n",
    "        ifelse(info$age < 60, \"A\", \"AM\")\n",
    "    )\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "info_p_AD <- info\n",
    "# Se aplica la función scale() para asegurar los prerrequisitos del modelo (media 0 y desviación estándar 1)\n",
    "# Se aplica para las columnas numéricas\n",
    "info_p_AD[, sapply(info_p_AD, is.numeric)] <- scale(info[, sapply(info, is.numeric)])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# A cada ocupación diferente se le asigna un número\n",
    "info_p_AD$occupation_as_n <- as.numeric(factor(info_p_AD$occupation))\n",
    "info_p_AD$occupation_as_n <- scale(info_p_AD$occupation_as_n)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>\n",
       ".dl-inline {width: auto; margin:0; padding: 0}\n",
       ".dl-inline>dt, .dl-inline>dd {float: none; width: auto; display: inline-block}\n",
       ".dl-inline>dt::after {content: \":\\0020\"; padding-right: .5ex}\n",
       ".dl-inline>dt:not(:first-of-type) {padding-left: .5ex}\n",
       "</style><dl class=dl-inline><dt>userid</dt><dd>-3.27870846791867e-17</dd><dt>age</dt><dd>-1.17503262979032e-16</dd><dt>itemid</dt><dd>4.18349770515795e-17</dd><dt>rate</dt><dd>-1.9700951807422e-16</dd><dt>timestamp</dt><dd>-8.01597080054596e-15</dd><dt>unknown</dt><dd>4.5544469617128e-18</dd><dt>Action</dt><dd>1.19982225110088e-17</dd><dt>Adventure</dt><dd>-3.6888116259487e-17</dd><dt>Animation</dt><dd>2.44849336660836e-17</dd><dt>Children</dt><dd>2.28238570621278e-17</dd><dt>Comedy</dt><dd>6.62573143052347e-17</dd><dt>Crime</dt><dd>1.76850867934439e-17</dd><dt>Documentary</dt><dd>-1.41769720420591e-18</dd><dt>Drama</dt><dd>-2.54123555965685e-17</dd><dt>Fantasy</dt><dd>-4.5409517617466e-19</dd><dt>FilmNoir</dt><dd>-4.97475758504118e-18</dd><dt>Horror</dt><dd>1.20070910137292e-17</dd><dt>Musical</dt><dd>-1.30812811212529e-17</dd><dt>Mystery</dt><dd>3.40904434130602e-17</dd><dt>Romance</dt><dd>-2.12063976405205e-17</dd><dt>SciFi</dt><dd>5.00228547820059e-18</dd><dt>Thriller</dt><dd>1.48351263998953e-17</dd><dt>War</dt><dd>1.58990165867803e-17</dd><dt>Western</dt><dd>-2.13515682456582e-17</dd><dt>gender_as_bit</dt><dd>-9.7371580199887e-17</dd><dt>occupation_as_n</dt><dd>3.12009532793533e-19</dd></dl>\n"
      ],
      "text/latex": [
       "\\begin{description*}\n",
       "\\item[userid] -3.27870846791867e-17\n",
       "\\item[age] -1.17503262979032e-16\n",
       "\\item[itemid] 4.18349770515795e-17\n",
       "\\item[rate] -1.9700951807422e-16\n",
       "\\item[timestamp] -8.01597080054596e-15\n",
       "\\item[unknown] 4.5544469617128e-18\n",
       "\\item[Action] 1.19982225110088e-17\n",
       "\\item[Adventure] -3.6888116259487e-17\n",
       "\\item[Animation] 2.44849336660836e-17\n",
       "\\item[Children] 2.28238570621278e-17\n",
       "\\item[Comedy] 6.62573143052347e-17\n",
       "\\item[Crime] 1.76850867934439e-17\n",
       "\\item[Documentary] -1.41769720420591e-18\n",
       "\\item[Drama] -2.54123555965685e-17\n",
       "\\item[Fantasy] -4.5409517617466e-19\n",
       "\\item[FilmNoir] -4.97475758504118e-18\n",
       "\\item[Horror] 1.20070910137292e-17\n",
       "\\item[Musical] -1.30812811212529e-17\n",
       "\\item[Mystery] 3.40904434130602e-17\n",
       "\\item[Romance] -2.12063976405205e-17\n",
       "\\item[SciFi] 5.00228547820059e-18\n",
       "\\item[Thriller] 1.48351263998953e-17\n",
       "\\item[War] 1.58990165867803e-17\n",
       "\\item[Western] -2.13515682456582e-17\n",
       "\\item[gender\\textbackslash{}\\_as\\textbackslash{}\\_bit] -9.7371580199887e-17\n",
       "\\item[occupation\\textbackslash{}\\_as\\textbackslash{}\\_n] 3.12009532793533e-19\n",
       "\\end{description*}\n"
      ],
      "text/markdown": [
       "userid\n",
       ":   -3.27870846791867e-17age\n",
       ":   -1.17503262979032e-16itemid\n",
       ":   4.18349770515795e-17rate\n",
       ":   -1.9700951807422e-16timestamp\n",
       ":   -8.01597080054596e-15unknown\n",
       ":   4.5544469617128e-18Action\n",
       ":   1.19982225110088e-17Adventure\n",
       ":   -3.6888116259487e-17Animation\n",
       ":   2.44849336660836e-17Children\n",
       ":   2.28238570621278e-17Comedy\n",
       ":   6.62573143052347e-17Crime\n",
       ":   1.76850867934439e-17Documentary\n",
       ":   -1.41769720420591e-18Drama\n",
       ":   -2.54123555965685e-17Fantasy\n",
       ":   -4.5409517617466e-19FilmNoir\n",
       ":   -4.97475758504118e-18Horror\n",
       ":   1.20070910137292e-17Musical\n",
       ":   -1.30812811212529e-17Mystery\n",
       ":   3.40904434130602e-17Romance\n",
       ":   -2.12063976405205e-17SciFi\n",
       ":   5.00228547820059e-18Thriller\n",
       ":   1.48351263998953e-17War\n",
       ":   1.58990165867803e-17Western\n",
       ":   -2.13515682456582e-17gender_as_bit\n",
       ":   -9.7371580199887e-17occupation_as_n\n",
       ":   3.12009532793533e-19\n",
       "\n"
      ],
      "text/plain": [
       "         userid             age          itemid            rate       timestamp \n",
       "  -3.278708e-17   -1.175033e-16    4.183498e-17   -1.970095e-16   -8.015971e-15 \n",
       "        unknown          Action       Adventure       Animation        Children \n",
       "   4.554447e-18    1.199822e-17   -3.688812e-17    2.448493e-17    2.282386e-17 \n",
       "         Comedy           Crime     Documentary           Drama         Fantasy \n",
       "   6.625731e-17    1.768509e-17   -1.417697e-18   -2.541236e-17   -4.540952e-19 \n",
       "       FilmNoir          Horror         Musical         Mystery         Romance \n",
       "  -4.974758e-18    1.200709e-17   -1.308128e-17    3.409044e-17   -2.120640e-17 \n",
       "          SciFi        Thriller             War         Western   gender_as_bit \n",
       "   5.002285e-18    1.483513e-17    1.589902e-17   -2.135157e-17   -9.737158e-17 \n",
       "occupation_as_n \n",
       "   3.120095e-19 "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<style>\n",
       ".dl-inline {width: auto; margin:0; padding: 0}\n",
       ".dl-inline>dt, .dl-inline>dd {float: none; width: auto; display: inline-block}\n",
       ".dl-inline>dt::after {content: \":\\0020\"; padding-right: .5ex}\n",
       ".dl-inline>dt:not(:first-of-type) {padding-left: .5ex}\n",
       "</style><dl class=dl-inline><dt>userid</dt><dd>1</dd><dt>age</dt><dd>1</dd><dt>itemid</dt><dd>1</dd><dt>rate</dt><dd>1</dd><dt>timestamp</dt><dd>1</dd><dt>unknown</dt><dd>1</dd><dt>Action</dt><dd>1</dd><dt>Adventure</dt><dd>1</dd><dt>Animation</dt><dd>1</dd><dt>Children</dt><dd>1</dd><dt>Comedy</dt><dd>0.999999999999999</dd><dt>Crime</dt><dd>1</dd><dt>Documentary</dt><dd>1</dd><dt>Drama</dt><dd>1</dd><dt>Fantasy</dt><dd>1</dd><dt>FilmNoir</dt><dd>1</dd><dt>Horror</dt><dd>1</dd><dt>Musical</dt><dd>1</dd><dt>Mystery</dt><dd>1</dd><dt>Romance</dt><dd>1</dd><dt>SciFi</dt><dd>1</dd><dt>Thriller</dt><dd>1</dd><dt>War</dt><dd>1</dd><dt>Western</dt><dd>0.999999999999999</dd><dt>gender_as_bit</dt><dd>1</dd><dt>occupation_as_n</dt><dd>1</dd></dl>\n"
      ],
      "text/latex": [
       "\\begin{description*}\n",
       "\\item[userid] 1\n",
       "\\item[age] 1\n",
       "\\item[itemid] 1\n",
       "\\item[rate] 1\n",
       "\\item[timestamp] 1\n",
       "\\item[unknown] 1\n",
       "\\item[Action] 1\n",
       "\\item[Adventure] 1\n",
       "\\item[Animation] 1\n",
       "\\item[Children] 1\n",
       "\\item[Comedy] 0.999999999999999\n",
       "\\item[Crime] 1\n",
       "\\item[Documentary] 1\n",
       "\\item[Drama] 1\n",
       "\\item[Fantasy] 1\n",
       "\\item[FilmNoir] 1\n",
       "\\item[Horror] 1\n",
       "\\item[Musical] 1\n",
       "\\item[Mystery] 1\n",
       "\\item[Romance] 1\n",
       "\\item[SciFi] 1\n",
       "\\item[Thriller] 1\n",
       "\\item[War] 1\n",
       "\\item[Western] 0.999999999999999\n",
       "\\item[gender\\textbackslash{}\\_as\\textbackslash{}\\_bit] 1\n",
       "\\item[occupation\\textbackslash{}\\_as\\textbackslash{}\\_n] 1\n",
       "\\end{description*}\n"
      ],
      "text/markdown": [
       "userid\n",
       ":   1age\n",
       ":   1itemid\n",
       ":   1rate\n",
       ":   1timestamp\n",
       ":   1unknown\n",
       ":   1Action\n",
       ":   1Adventure\n",
       ":   1Animation\n",
       ":   1Children\n",
       ":   1Comedy\n",
       ":   0.999999999999999Crime\n",
       ":   1Documentary\n",
       ":   1Drama\n",
       ":   1Fantasy\n",
       ":   1FilmNoir\n",
       ":   1Horror\n",
       ":   1Musical\n",
       ":   1Mystery\n",
       ":   1Romance\n",
       ":   1SciFi\n",
       ":   1Thriller\n",
       ":   1War\n",
       ":   1Western\n",
       ":   0.999999999999999gender_as_bit\n",
       ":   1occupation_as_n\n",
       ":   1\n",
       "\n"
      ],
      "text/plain": [
       "         userid             age          itemid            rate       timestamp \n",
       "              1               1               1               1               1 \n",
       "        unknown          Action       Adventure       Animation        Children \n",
       "              1               1               1               1               1 \n",
       "         Comedy           Crime     Documentary           Drama         Fantasy \n",
       "              1               1               1               1               1 \n",
       "       FilmNoir          Horror         Musical         Mystery         Romance \n",
       "              1               1               1               1               1 \n",
       "          SciFi        Thriller             War         Western   gender_as_bit \n",
       "              1               1               1               1               1 \n",
       "occupation_as_n \n",
       "              1 "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Se verifica\n",
    "apply(info_p_AD[, sapply(info_p_AD, is.numeric)], 2, mean)\n",
    "apply(info_p_AD[, sapply(info_p_AD, is.numeric)], 2, sd)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se separan las muestras en entrenamiento y prueba\n",
    "separador_AD <- sample(1:nrow(info_p_AD), 0.8 * nrow(info_p_AD))\n",
    "info_entrenamiento_AD <- info_p_AD[separador_AD, ]\n",
    "info_prueba_AD <- info_p_AD[-separador_AD, ]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se crea el modelo de análisis discriminante, seleccionando 5 géneros de películas como referencia\n",
    "modelo_AD <- lda(age_group ~ rate + occupation_as_n + gender_as_bit + Action + Children + Crime + Romance + SciFi,\n",
    "    data = info_entrenamiento_AD\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se obtienen 3 registros de prueba, sin la columna de grupo etario\n",
    "reg_prueba_AD <- info_prueba_AD[1:3, ]\n",
    "# Se obtienen las probabilidades de que el usuario pertenezca a cada grupo etario (la predicción)\n",
    "prediccion_AD <- predict(modelo_AD, reg_prueba_AD)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"dataframe\">\n",
       "<caption>A data.frame: 3 × 34</caption>\n",
       "<thead>\n",
       "\t<tr><th></th><th scope=col>userid</th><th scope=col>age</th><th scope=col>gender</th><th scope=col>occupation</th><th scope=col>zcode</th><th scope=col>itemid</th><th scope=col>rate</th><th scope=col>timestamp</th><th scope=col>title</th><th scope=col>rdate</th><th scope=col>⋯</th><th scope=col>Musical</th><th scope=col>Mystery</th><th scope=col>Romance</th><th scope=col>SciFi</th><th scope=col>Thriller</th><th scope=col>War</th><th scope=col>Western</th><th scope=col>gender_as_bit</th><th scope=col>age_group</th><th scope=col>occupation_as_n</th></tr>\n",
       "\t<tr><th></th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>⋯</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;dbl[,1]&gt;</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><th scope=row>1</th><td>-1.730907</td><td>-0.7757625</td><td>M</td><td>technician</td><td>85711</td><td>-1.1019708</td><td>0.4176522</td><td>-0.9331148</td><td>Three Colors: White (1994)     </td><td>01-Jan-1994</td><td>⋯</td><td>-0.2283016</td><td>-0.2352716</td><td>-0.4915609</td><td>-0.3819263</td><td>-0.5291012</td><td>-0.3220673</td><td>-0.137441</td><td>-0.5887415</td><td>A</td><td>1.196413</td></tr>\n",
       "\t<tr><th scope=row>12</th><td>-1.730907</td><td>-0.7757625</td><td>M</td><td>technician</td><td>85711</td><td>-0.6152695</td><td>0.4176522</td><td>-0.8711805</td><td>Star Trek: First Contact (1996)</td><td>22-Nov-1996</td><td>⋯</td><td>-0.2283016</td><td>-0.2352716</td><td>-0.4915609</td><td> 2.6182802</td><td>-0.5291012</td><td>-0.3220673</td><td>-0.137441</td><td>-0.5887415</td><td>A</td><td>1.196413</td></tr>\n",
       "\t<tr><th scope=row>13</th><td>-1.730907</td><td>-0.7757625</td><td>M</td><td>technician</td><td>85711</td><td>-0.5215568</td><td>1.3060091</td><td>-1.6023787</td><td>Pillow Book, The (1995)        </td><td>13-Jun-1997</td><td>⋯</td><td>-0.2283016</td><td>-0.2352716</td><td> 2.0343158</td><td>-0.3819263</td><td>-0.5291012</td><td>-0.3220673</td><td>-0.137441</td><td>-0.5887415</td><td>A</td><td>1.196413</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "A data.frame: 3 × 34\n",
       "\\begin{tabular}{r|lllllllllllllllllllll}\n",
       "  & userid & age & gender & occupation & zcode & itemid & rate & timestamp & title & rdate & ⋯ & Musical & Mystery & Romance & SciFi & Thriller & War & Western & gender\\_as\\_bit & age\\_group & occupation\\_as\\_n\\\\\n",
       "  & <dbl> & <dbl> & <chr> & <chr> & <chr> & <dbl> & <dbl> & <dbl> & <chr> & <chr> & ⋯ & <dbl> & <dbl> & <dbl> & <dbl> & <dbl> & <dbl> & <dbl> & <dbl> & <chr> & <dbl{[},1{]}>\\\\\n",
       "\\hline\n",
       "\t1 & -1.730907 & -0.7757625 & M & technician & 85711 & -1.1019708 & 0.4176522 & -0.9331148 & Three Colors: White (1994)      & 01-Jan-1994 & ⋯ & -0.2283016 & -0.2352716 & -0.4915609 & -0.3819263 & -0.5291012 & -0.3220673 & -0.137441 & -0.5887415 & A & 1.196413\\\\\n",
       "\t12 & -1.730907 & -0.7757625 & M & technician & 85711 & -0.6152695 & 0.4176522 & -0.8711805 & Star Trek: First Contact (1996) & 22-Nov-1996 & ⋯ & -0.2283016 & -0.2352716 & -0.4915609 &  2.6182802 & -0.5291012 & -0.3220673 & -0.137441 & -0.5887415 & A & 1.196413\\\\\n",
       "\t13 & -1.730907 & -0.7757625 & M & technician & 85711 & -0.5215568 & 1.3060091 & -1.6023787 & Pillow Book, The (1995)         & 13-Jun-1997 & ⋯ & -0.2283016 & -0.2352716 &  2.0343158 & -0.3819263 & -0.5291012 & -0.3220673 & -0.137441 & -0.5887415 & A & 1.196413\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "A data.frame: 3 × 34\n",
       "\n",
       "| <!--/--> | userid &lt;dbl&gt; | age &lt;dbl&gt; | gender &lt;chr&gt; | occupation &lt;chr&gt; | zcode &lt;chr&gt; | itemid &lt;dbl&gt; | rate &lt;dbl&gt; | timestamp &lt;dbl&gt; | title &lt;chr&gt; | rdate &lt;chr&gt; | ⋯ ⋯ | Musical &lt;dbl&gt; | Mystery &lt;dbl&gt; | Romance &lt;dbl&gt; | SciFi &lt;dbl&gt; | Thriller &lt;dbl&gt; | War &lt;dbl&gt; | Western &lt;dbl&gt; | gender_as_bit &lt;dbl&gt; | age_group &lt;chr&gt; | occupation_as_n &lt;dbl[,1]&gt; |\n",
       "|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n",
       "| 1 | -1.730907 | -0.7757625 | M | technician | 85711 | -1.1019708 | 0.4176522 | -0.9331148 | Three Colors: White (1994)      | 01-Jan-1994 | ⋯ | -0.2283016 | -0.2352716 | -0.4915609 | -0.3819263 | -0.5291012 | -0.3220673 | -0.137441 | -0.5887415 | A | 1.196413 |\n",
       "| 12 | -1.730907 | -0.7757625 | M | technician | 85711 | -0.6152695 | 0.4176522 | -0.8711805 | Star Trek: First Contact (1996) | 22-Nov-1996 | ⋯ | -0.2283016 | -0.2352716 | -0.4915609 |  2.6182802 | -0.5291012 | -0.3220673 | -0.137441 | -0.5887415 | A | 1.196413 |\n",
       "| 13 | -1.730907 | -0.7757625 | M | technician | 85711 | -0.5215568 | 1.3060091 | -1.6023787 | Pillow Book, The (1995)         | 13-Jun-1997 | ⋯ | -0.2283016 | -0.2352716 |  2.0343158 | -0.3819263 | -0.5291012 | -0.3220673 | -0.137441 | -0.5887415 | A | 1.196413 |\n",
       "\n"
      ],
      "text/plain": [
       "   userid    age        gender occupation zcode itemid     rate      timestamp \n",
       "1  -1.730907 -0.7757625 M      technician 85711 -1.1019708 0.4176522 -0.9331148\n",
       "12 -1.730907 -0.7757625 M      technician 85711 -0.6152695 0.4176522 -0.8711805\n",
       "13 -1.730907 -0.7757625 M      technician 85711 -0.5215568 1.3060091 -1.6023787\n",
       "   title                           rdate       ⋯ Musical    Mystery   \n",
       "1  Three Colors: White (1994)      01-Jan-1994 ⋯ -0.2283016 -0.2352716\n",
       "12 Star Trek: First Contact (1996) 22-Nov-1996 ⋯ -0.2283016 -0.2352716\n",
       "13 Pillow Book, The (1995)         13-Jun-1997 ⋯ -0.2283016 -0.2352716\n",
       "   Romance    SciFi      Thriller   War        Western   gender_as_bit\n",
       "1  -0.4915609 -0.3819263 -0.5291012 -0.3220673 -0.137441 -0.5887415   \n",
       "12 -0.4915609  2.6182802 -0.5291012 -0.3220673 -0.137441 -0.5887415   \n",
       "13  2.0343158 -0.3819263 -0.5291012 -0.3220673 -0.137441 -0.5887415   \n",
       "   age_group occupation_as_n\n",
       "1  A         1.196413       \n",
       "12 A         1.196413       \n",
       "13 A         1.196413       "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<dl>\n",
       "\t<dt>$class</dt>\n",
       "\t\t<dd><style>\n",
       ".list-inline {list-style: none; margin:0; padding: 0}\n",
       ".list-inline>li {display: inline-block}\n",
       ".list-inline>li:not(:last-child)::after {content: \"\\00b7\"; padding: 0 .5ex}\n",
       "</style>\n",
       "<ol class=list-inline><li>A</li><li>A</li><li>A</li></ol>\n",
       "\n",
       "<details>\n",
       "\t<summary style=display:list-item;cursor:pointer>\n",
       "\t\t<strong>Levels</strong>:\n",
       "\t</summary>\n",
       "\t<style>\n",
       "\t.list-inline {list-style: none; margin:0; padding: 0}\n",
       "\t.list-inline>li {display: inline-block}\n",
       "\t.list-inline>li:not(:last-child)::after {content: \"\\00b7\"; padding: 0 .5ex}\n",
       "\t</style>\n",
       "\t<ol class=list-inline><li>'a'</li><li>'A'</li><li>'AM'</li><li>'ñ'</li></ol>\n",
       "</details></dd>\n",
       "\t<dt>$posterior</dt>\n",
       "\t\t<dd><table class=\"dataframe\">\n",
       "<caption>A matrix: 3 × 4 of type dbl</caption>\n",
       "<thead>\n",
       "\t<tr><th></th><th scope=col>a</th><th scope=col>A</th><th scope=col>AM</th><th scope=col>ñ</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><th scope=row>1</th><td>0.03927417</td><td>0.9175351</td><td>0.04252513</td><td>0.0006656147</td></tr>\n",
       "\t<tr><th scope=row>12</th><td>0.04562316</td><td>0.9331500</td><td>0.01907439</td><td>0.0021524867</td></tr>\n",
       "\t<tr><th scope=row>13</th><td>0.03610820</td><td>0.9117856</td><td>0.05023596</td><td>0.0018702482</td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</dd>\n",
       "\t<dt>$x</dt>\n",
       "\t\t<dd><table class=\"dataframe\">\n",
       "<caption>A matrix: 3 × 3 of type dbl</caption>\n",
       "<thead>\n",
       "\t<tr><th></th><th scope=col>LD1</th><th scope=col>LD2</th><th scope=col>LD3</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><th scope=row>1</th><td>-0.8632885</td><td> 1.18641852</td><td>-0.5662896</td></tr>\n",
       "\t<tr><th scope=row>12</th><td>-1.1508896</td><td>-0.07824143</td><td> 1.9442907</td></tr>\n",
       "\t<tr><th scope=row>13</th><td>-0.7777013</td><td> 1.48606838</td><td> 0.7510413</td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</dd>\n",
       "</dl>\n"
      ],
      "text/latex": [
       "\\begin{description}\n",
       "\\item[\\$class] \\begin{enumerate*}\n",
       "\\item A\n",
       "\\item A\n",
       "\\item A\n",
       "\\end{enumerate*}\n",
       "\n",
       "\\emph{Levels}: \\begin{enumerate*}\n",
       "\\item 'a'\n",
       "\\item 'A'\n",
       "\\item 'AM'\n",
       "\\item 'ñ'\n",
       "\\end{enumerate*}\n",
       "\n",
       "\\item[\\$posterior] A matrix: 3 × 4 of type dbl\n",
       "\\begin{tabular}{r|llll}\n",
       "  & a & A & AM & ñ\\\\\n",
       "\\hline\n",
       "\t1 & 0.03927417 & 0.9175351 & 0.04252513 & 0.0006656147\\\\\n",
       "\t12 & 0.04562316 & 0.9331500 & 0.01907439 & 0.0021524867\\\\\n",
       "\t13 & 0.03610820 & 0.9117856 & 0.05023596 & 0.0018702482\\\\\n",
       "\\end{tabular}\n",
       "\n",
       "\\item[\\$x] A matrix: 3 × 3 of type dbl\n",
       "\\begin{tabular}{r|lll}\n",
       "  & LD1 & LD2 & LD3\\\\\n",
       "\\hline\n",
       "\t1 & -0.8632885 &  1.18641852 & -0.5662896\\\\\n",
       "\t12 & -1.1508896 & -0.07824143 &  1.9442907\\\\\n",
       "\t13 & -0.7777013 &  1.48606838 &  0.7510413\\\\\n",
       "\\end{tabular}\n",
       "\n",
       "\\end{description}\n"
      ],
      "text/markdown": [
       "$class\n",
       ":   1. A\n",
       "2. A\n",
       "3. A\n",
       "\n",
       "\n",
       "\n",
       "**Levels**: 1. 'a'\n",
       "2. 'A'\n",
       "3. 'AM'\n",
       "4. 'ñ'\n",
       "\n",
       "\n",
       "\n",
       "$posterior\n",
       ":   \n",
       "A matrix: 3 × 4 of type dbl\n",
       "\n",
       "| <!--/--> | a | A | AM | ñ |\n",
       "|---|---|---|---|---|\n",
       "| 1 | 0.03927417 | 0.9175351 | 0.04252513 | 0.0006656147 |\n",
       "| 12 | 0.04562316 | 0.9331500 | 0.01907439 | 0.0021524867 |\n",
       "| 13 | 0.03610820 | 0.9117856 | 0.05023596 | 0.0018702482 |\n",
       "\n",
       "\n",
       "$x\n",
       ":   \n",
       "A matrix: 3 × 3 of type dbl\n",
       "\n",
       "| <!--/--> | LD1 | LD2 | LD3 |\n",
       "|---|---|---|---|\n",
       "| 1 | -0.8632885 |  1.18641852 | -0.5662896 |\n",
       "| 12 | -1.1508896 | -0.07824143 |  1.9442907 |\n",
       "| 13 | -0.7777013 |  1.48606838 |  0.7510413 |\n",
       "\n",
       "\n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "$class\n",
       "[1] A A A\n",
       "Levels: a A AM ñ\n",
       "\n",
       "$posterior\n",
       "            a         A         AM            ñ\n",
       "1  0.03927417 0.9175351 0.04252513 0.0006656147\n",
       "12 0.04562316 0.9331500 0.01907439 0.0021524867\n",
       "13 0.03610820 0.9117856 0.05023596 0.0018702482\n",
       "\n",
       "$x\n",
       "          LD1         LD2        LD3\n",
       "1  -0.8632885  1.18641852 -0.5662896\n",
       "12 -1.1508896 -0.07824143  1.9442907\n",
       "13 -0.7777013  1.48606838  0.7510413\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "reg_prueba_AD\n",
    "prediccion_AD\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "prediccion_AD <- predict(modelo_AD, info_prueba_AD)$x\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 420,
       "width": 420
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_AD <- cbind(info_entrenamiento_AD, prediccion_AD)\n",
    "# Realizamos una predicción para todos los datos de prueba y observamos la separación de los grupos\n",
    "ggplot(plot_AD, aes(LD1, LD2)) +\n",
    "  geom_point(aes(color = age_group))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A través de este gráfico se puede concluir que, tal como se mecionaba con anterioridad, los datos son bastante heterogeneos, por lo que realizar una separación de los grupos etarios presenta un desafío para el modelo de análisis discriminante, debido a que muchos usuarios de diferentes edades comparten gustos de cine.\n",
    "\n",
    "Para realizar esta separación con mayor éxito, debe existir una descripción más exhaustiva del usuario que califica la película, así como posiblemente emplear más columnas de género para aumentar la precisión del modelo de AD."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"dataframe\">\n",
       "<caption>A data.frame: 6 × 3</caption>\n",
       "<thead>\n",
       "\t<tr><th></th><th scope=col>age</th><th scope=col>rate</th><th scope=col>occupation</th></tr>\n",
       "\t<tr><th></th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;chr&gt;</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><th scope=row>1</th><td>24</td><td>4</td><td>technician</td></tr>\n",
       "\t<tr><th scope=row>2</th><td>24</td><td>3</td><td>technician</td></tr>\n",
       "\t<tr><th scope=row>3</th><td>24</td><td>4</td><td>technician</td></tr>\n",
       "\t<tr><th scope=row>4</th><td>24</td><td>4</td><td>technician</td></tr>\n",
       "\t<tr><th scope=row>5</th><td>24</td><td>4</td><td>technician</td></tr>\n",
       "\t<tr><th scope=row>6</th><td>24</td><td>5</td><td>technician</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "A data.frame: 6 × 3\n",
       "\\begin{tabular}{r|lll}\n",
       "  & age & rate & occupation\\\\\n",
       "  & <int> & <int> & <chr>\\\\\n",
       "\\hline\n",
       "\t1 & 24 & 4 & technician\\\\\n",
       "\t2 & 24 & 3 & technician\\\\\n",
       "\t3 & 24 & 4 & technician\\\\\n",
       "\t4 & 24 & 4 & technician\\\\\n",
       "\t5 & 24 & 4 & technician\\\\\n",
       "\t6 & 24 & 5 & technician\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "A data.frame: 6 × 3\n",
       "\n",
       "| <!--/--> | age &lt;int&gt; | rate &lt;int&gt; | occupation &lt;chr&gt; |\n",
       "|---|---|---|---|\n",
       "| 1 | 24 | 4 | technician |\n",
       "| 2 | 24 | 3 | technician |\n",
       "| 3 | 24 | 4 | technician |\n",
       "| 4 | 24 | 4 | technician |\n",
       "| 5 | 24 | 4 | technician |\n",
       "| 6 | 24 | 5 | technician |\n",
       "\n"
      ],
      "text/plain": [
       "  age rate occupation\n",
       "1 24  4    technician\n",
       "2 24  3    technician\n",
       "3 24  4    technician\n",
       "4 24  4    technician\n",
       "5 24  4    technician\n",
       "6 24  5    technician"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Planteamiento para modelo k-vecinos\n",
    "# Se desea saber la ocupación de un usuario a partir de su edad y calificaciones\n",
    "\n",
    "# Se recuperarán únicamentélas columnas de edad, calificaciones y ocupación, ya que\n",
    "# el algoritmo de k-vecinos, en primera instancia, sólo trabaja con datos numéricos.\n",
    "info_p_KV <- info[, c(\"age\", \"rate\", \"occupation\")]\n",
    "head(info_p_KV)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "info_p_KV <- unique(info_p_KV)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "2067"
      ],
      "text/latex": [
       "2067"
      ],
      "text/markdown": [
       "2067"
      ],
      "text/plain": [
       "[1] 2067"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "nrow(info_p_KV)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 420,
       "width": 420
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ggplot(info_p_KV, aes(x = age, y = rate, color = occupation)) +\n",
    "    geom_point() +\n",
    "    labs(title = \"Relación de edad y calificaciones para cada ocupación\", x = \"Edad\", y = \"Calificaciones\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Con ayuda de este gráfico se puede visualizar cómo las ocupaciones de distribuyen de manera similar en el eje vertical (indicando que, posiblemente, las calificaciones de un usuario están poco relacionadas con su ocupación), pero presentan una vasta heterogeneidad en el eje horizontal. Podemos ver cómo múltiples colores (ocupaciones) se aglomeran en ciertos rangos etarios, por lo que se espera que esta variable sea de gran injerencia al realizar una predicción."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se declara la función para normalizar los datos, que permite deshacerse de diferencias\n",
    "# de magnitud entre los datos y nos permite graficarlos en un mismo plano.\n",
    "normalizar <-\n",
    "    function(x) {\n",
    "        (x - min(x)) / (max(x) - min(x))\n",
    "    }\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"dataframe\">\n",
       "<caption>A data.frame: 6 × 2</caption>\n",
       "<thead>\n",
       "\t<tr><th></th><th scope=col>age</th><th scope=col>rate</th></tr>\n",
       "\t<tr><th></th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><th scope=row>1</th><td>0.2575758</td><td>0.75</td></tr>\n",
       "\t<tr><th scope=row>2</th><td>0.2575758</td><td>0.50</td></tr>\n",
       "\t<tr><th scope=row>3</th><td>0.2575758</td><td>1.00</td></tr>\n",
       "\t<tr><th scope=row>4</th><td>0.2575758</td><td>0.25</td></tr>\n",
       "\t<tr><th scope=row>5</th><td>0.2575758</td><td>0.00</td></tr>\n",
       "\t<tr><th scope=row>6</th><td>0.6969697</td><td>0.75</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "A data.frame: 6 × 2\n",
       "\\begin{tabular}{r|ll}\n",
       "  & age & rate\\\\\n",
       "  & <dbl> & <dbl>\\\\\n",
       "\\hline\n",
       "\t1 & 0.2575758 & 0.75\\\\\n",
       "\t2 & 0.2575758 & 0.50\\\\\n",
       "\t3 & 0.2575758 & 1.00\\\\\n",
       "\t4 & 0.2575758 & 0.25\\\\\n",
       "\t5 & 0.2575758 & 0.00\\\\\n",
       "\t6 & 0.6969697 & 0.75\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "A data.frame: 6 × 2\n",
       "\n",
       "| <!--/--> | age &lt;dbl&gt; | rate &lt;dbl&gt; |\n",
       "|---|---|---|\n",
       "| 1 | 0.2575758 | 0.75 |\n",
       "| 2 | 0.2575758 | 0.50 |\n",
       "| 3 | 0.2575758 | 1.00 |\n",
       "| 4 | 0.2575758 | 0.25 |\n",
       "| 5 | 0.2575758 | 0.00 |\n",
       "| 6 | 0.6969697 | 0.75 |\n",
       "\n"
      ],
      "text/plain": [
       "  age       rate\n",
       "1 0.2575758 0.75\n",
       "2 0.2575758 0.50\n",
       "3 0.2575758 1.00\n",
       "4 0.2575758 0.25\n",
       "5 0.2575758 0.00\n",
       "6 0.6969697 0.75"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "info_p_KV_norm <- as.data.frame(lapply(info_p_KV[, c(\"age\", \"rate\")], normalizar))\n",
    "head(info_p_KV_norm)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Separación de los datos en entrenamiento y prueba\n",
    "separador_KV <- sample(1:nrow(info_p_KV_norm), 0.8 * nrow(info_p_KV_norm))\n",
    "info_entrenamiento_KV <- info_p_KV_norm[separador_KV, ]\n",
    "info_prueba_KV <- info_p_KV_norm[-separador_KV, ]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "# Se crea el modelo de k-vecinos\n",
    "predicciones <- knn(info_entrenamiento_KV, # Datos de entrenamiento\n",
    "    info_prueba_KV, # Datos de prueba\n",
    "    info_p_KV[separador_KV, 3], # Etiquetas bajo las cuales se clasificarán los datos de entrenamiento\n",
    "    k = 7\n",
    ") # Número de vecinos\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "               \n",
       "predicciones    administrator artist doctor educator engineer entertainment\n",
       "  administrator             0      3      0        1        3             0\n",
       "  artist                    2      0      1        1        0             0\n",
       "  doctor                    0      0      0        0        1             0\n",
       "  educator                  3      4      0        0        7             2\n",
       "  engineer                  4      2      0        3        1             1\n",
       "  entertainment             0      0      0        0        0             0\n",
       "  executive                 4      0      0        4        0             1\n",
       "  healthcare                0      0      0        0        0             0\n",
       "  homemaker                 0      0      0        0        0             1\n",
       "  lawyer                    2      0      1        0        2             1\n",
       "  librarian                 2      2      0        3        2             1\n",
       "  marketing                 1      1      0        0        1             2\n",
       "  none                      0      0      0        0        0             0\n",
       "  other                     4      1      0        5        3             1\n",
       "  programmer                3      0      0        2        3             1\n",
       "  retired                   0      0      1        0        0             0\n",
       "  salesman                  0      0      0        0        1             0\n",
       "  scientist                 1      2      0        2        2             2\n",
       "  student                   1      5      2        1        1             3\n",
       "  technician                1      0      0        0        1             1\n",
       "  writer                    1      0      1        3        3             0\n",
       "               \n",
       "predicciones    executive healthcare homemaker lawyer librarian marketing none\n",
       "  administrator         5          2         0      1         0         1    1\n",
       "  artist                1          0         0      0         0         1    0\n",
       "  doctor                1          0         1      0         0         0    0\n",
       "  educator              2          4         2      0         9         1    1\n",
       "  engineer              1          2         1      0         3         3    0\n",
       "  entertainment         0          0         0      0         2         1    0\n",
       "  executive             0          3         0      1         4         1    2\n",
       "  healthcare            0          0         0      0         0         0    0\n",
       "  homemaker             0          0         0      0         0         0    0\n",
       "  lawyer                0          0         1      0         1         2    1\n",
       "  librarian             1          0         2      1         0         0    1\n",
       "  marketing             0          1         0      2         0         0    0\n",
       "  none                  0          0         0      0         1         0    0\n",
       "  other                 1          0         0      2         2         1    0\n",
       "  programmer            1          0         0      0         1         1    1\n",
       "  retired               2          0         0      0         0         0    0\n",
       "  salesman              0          0         0      0         0         0    0\n",
       "  scientist             2          1         0      0         1         1    1\n",
       "  student               1          1         2      2         0         1    2\n",
       "  technician            0          2         0      2         1         0    0\n",
       "  writer                1          2         1      1         0         0    0\n",
       "               \n",
       "predicciones    other programmer retired salesman scientist student technician\n",
       "  administrator     2          3       0        1         0       2          2\n",
       "  artist            1          1       0        0         2       0          1\n",
       "  doctor            0          0       1        0         0       0          0\n",
       "  educator          6          9       1        1         1       0          1\n",
       "  engineer          3          1       1        1         1       1          2\n",
       "  entertainment     1          1       0        0         1       0          0\n",
       "  executive         1          1       1        0         4       1          1\n",
       "  healthcare        2          0       0        0         0       0          0\n",
       "  homemaker         0          0       0        0         1       0          0\n",
       "  lawyer            2          0       0        2         0       0          0\n",
       "  librarian         6          1       0        0         4       0          3\n",
       "  marketing         1          0       0        1         0       2          1\n",
       "  none              1          0       0        1         0       1          1\n",
       "  other             0          4       2        2         5       1          2\n",
       "  programmer        5          0       0        1         3       3          1\n",
       "  retired           0          0       0        0         0       0          0\n",
       "  salesman          0          1       0        0         0       1          0\n",
       "  scientist         2          1       0        0         1       2          1\n",
       "  student           5          3       0        3         0       3          1\n",
       "  technician        0          0       0        0         0       2          0\n",
       "  writer            5          0       0        0         0       2          1\n",
       "               \n",
       "predicciones    writer\n",
       "  administrator      4\n",
       "  artist             1\n",
       "  doctor             0\n",
       "  educator           4\n",
       "  engineer           3\n",
       "  entertainment      0\n",
       "  executive          1\n",
       "  healthcare         2\n",
       "  homemaker          0\n",
       "  lawyer             1\n",
       "  librarian          0\n",
       "  marketing          1\n",
       "  none               0\n",
       "  other              4\n",
       "  programmer         3\n",
       "  retired            0\n",
       "  salesman           0\n",
       "  scientist          0\n",
       "  student            4\n",
       "  technician         0\n",
       "  writer             0"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Se crea una matriz de confusión para ver la precisión del algoritmo\n",
    "tab <- table(predicciones, info_p_KV[-separador_KV, 3])\n",
    "tab\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "1.20772946859903"
      ],
      "text/latex": [
       "1.20772946859903"
      ],
      "text/markdown": [
       "1.20772946859903"
      ],
      "text/plain": [
       "[1] 1.207729"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# En esta función se divide el número de predicciones correctas entre el número total de predicciones, lo que nos dice\n",
    "# qué tan preciso es el modelo.\n",
    "precision(tab)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nuevamente, la diversidad de la información (y posiblemente la falta de esta) empobrecen el rendimiento de nuestro modelo, pues al igual que otras variables de clasificación (a excepción del género de la película), dicen poco de cómo un usuario se comporta al calificar.\n",
    "\n",
    "Finalmente, sería importante revisar la vigencia de estos datos, y/o enriquecer los planteamientos con más detalles, pues a partir de pocas variables es difícil ser certero en las predicciones."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "R",
   "language": "R",
   "name": "ir"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "4.2.2"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "26de051ba29f2982a8de78e945f0abaf191376122a1563185a90213a26c5da77"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}