Output_Age_ion-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "library(MASS)\n",
    "library(plyr)\n",
    "library(dplyr)\n",
    "library(readr)\n",
    "library(ggplot2)\n",
    "library(reshape2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "options(repr.plot.height=5)\n",
    "options(repr.matrix.max.rows=20)\n",
    "options(repr.matrix.max.columns=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load Lenskit 1M"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Warning message in lenskit.meta$age[as.numeric(lenskit.meta$age) >= 18 & as.numeric(lenskit.meta$age) < :\n",
      "“NAs introduced by coercion”Warning message in lenskit.meta$age[as.numeric(lenskit.meta$age) >= 18 & as.numeric(lenskit.meta$age) < :\n",
      "“NAs introduced by coercion”Warning message in lenskit.meta$age[as.numeric(lenskit.meta$age) >= 25 & as.numeric(lenskit.meta$age) < :\n",
      "“NAs introduced by coercion”Warning message in lenskit.meta$age[as.numeric(lenskit.meta$age) >= 25 & as.numeric(lenskit.meta$age) < :\n",
      "“NAs introduced by coercion”Warning message in lenskit.meta$age[as.numeric(lenskit.meta$age) >= 35 & as.numeric(lenskit.meta$age) < :\n",
      "“NAs introduced by coercion”Warning message in lenskit.meta$age[as.numeric(lenskit.meta$age) >= 35 & as.numeric(lenskit.meta$age) < :\n",
      "“NAs introduced by coercion”Warning message in lenskit.meta$age[as.numeric(lenskit.meta$age) >= 45 & as.numeric(lenskit.meta$age) < :\n",
      "“NAs introduced by coercion”Warning message in lenskit.meta$age[as.numeric(lenskit.meta$age) >= 45 & as.numeric(lenskit.meta$age) < :\n",
      "“NAs introduced by coercion”Warning message in lenskit.meta$age[as.numeric(lenskit.meta$age) >= 56] <- \"[56+]\":\n",
      "“NAs introduced by coercion”"
     ]
    }
   ],
   "source": [
    "lenskit.meta = read_delim(\"data/lk1m/users.dat\", delim=\":\",\n",
    "                   col_names=c(\"user\", \"gender\", \"age\", \"occupation\", \"zip\"),\n",
    "                   col_types=\"i_c_c_c_c\") %>%\n",
    "    mutate(gender=as.factor(gender),\n",
    "           \n",
    "          dataset = \"Lenskit-1m\")\n",
    "\n",
    "lenskit.meta$age[ as.numeric(lenskit.meta$age) < 18] <- \"[1-17]\"\n",
    "lenskit.meta$age[as.numeric(lenskit.meta$age) >= 18 & as.numeric(lenskit.meta$age) < 25] <- \"[18-24]\"\n",
    "lenskit.meta$age[as.numeric(lenskit.meta$age) >= 25 &  as.numeric(lenskit.meta$age) < 35] <- \"[25-34]\"\n",
    "lenskit.meta$age[as.numeric(lenskit.meta$age) >= 35 &  as.numeric(lenskit.meta$age) < 45] <- \"[35-44]\"\n",
    "lenskit.meta$age[as.numeric(lenskit.meta$age) >= 45 & as.numeric(lenskit.meta$age) < 56] <- \"[45-55]\"\n",
    "lenskit.meta$age[as.numeric(lenskit.meta$age) >= 56 ] <- \"[56+]\"\n",
    "lenskit.meta$age = as.factor(lenskit.meta$age)\n",
    "#lenskit.meta\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Parsed with column specification:\n",
      "cols(\n",
      "  DataSet = col_character(),\n",
      "  Partition = col_integer(),\n",
      "  Algorithm = col_character(),\n",
      "  User = col_integer(),\n",
      "  TestTime = col_double(),\n",
      "  TopN.nDCG = col_double(),\n",
      "  Rank = col_integer(),\n",
      "  RecipRank = col_double(),\n",
      "  AvgPrec = col_double()\n",
      ")\n",
      "Joining, by = \"user\"\n"
     ]
    }
   ],
   "source": [
    "user.results.lenskit1m = read_csv(\"data/lk1m/eval-users.csv\", guess_max=10000) %>%\n",
    "    rename(user=User) %>%\n",
    "    inner_join(users.meta %>% rename(user=id) %>% select(-key))\n",
    "user.results.lenskit1m$DataSet=\"Lenskit1M\"\n",
    "#user.results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load Last.FM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Parsed with column specification:\n",
      "cols(\n",
      "  id = col_integer(),\n",
      "  key = col_character(),\n",
      "  gender = col_character(),\n",
      "  age = col_integer()\n",
      ")\n",
      "Warning message in usersLastfm$age[as.numeric(usersLastfm$age) >= 18 & as.numeric(usersLastfm$age) < :\n",
      "“NAs introduced by coercion”Warning message in usersLastfm$age[as.numeric(usersLastfm$age) >= 18 & as.numeric(usersLastfm$age) < :\n",
      "“NAs introduced by coercion”Warning message in usersLastfm$age[as.numeric(usersLastfm$age) >= 25 & as.numeric(usersLastfm$age) < :\n",
      "“NAs introduced by coercion”Warning message in usersLastfm$age[as.numeric(usersLastfm$age) >= 25 & as.numeric(usersLastfm$age) < :\n",
      "“NAs introduced by coercion”Warning message in usersLastfm$age[as.numeric(usersLastfm$age) >= 35 & as.numeric(usersLastfm$age) < :\n",
      "“NAs introduced by coercion”Warning message in usersLastfm$age[as.numeric(usersLastfm$age) >= 35 & as.numeric(usersLastfm$age) < :\n",
      "“NAs introduced by coercion”Warning message in usersLastfm$age[as.numeric(usersLastfm$age) >= 45 & as.numeric(usersLastfm$age) < :\n",
      "“NAs introduced by coercion”Warning message in usersLastfm$age[as.numeric(usersLastfm$age) >= 45 & as.numeric(usersLastfm$age) < :\n",
      "“NAs introduced by coercion”Warning message in usersLastfm$age[as.numeric(usersLastfm$age) >= 56] <- \"[56+]\":\n",
      "“NAs introduced by coercion”"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>id</th><th scope=col>key</th><th scope=col>gender</th><th scope=col>age</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td> 1                                      </td><td>00000c289a1829a808ac09c00daf10bc3c4e223b</td><td>F                                       </td><td>[18-24]                                 </td></tr>\n",
       "\t<tr><td> 2                                      </td><td>00001411dc427966b17297bf4d69e7e193135d89</td><td>F                                       </td><td>NA                                      </td></tr>\n",
       "\t<tr><td> 3                                      </td><td>00004d2ac9316e22dc007ab2243d6fcb239e707d</td><td>NA                                      </td><td>NA                                      </td></tr>\n",
       "\t<tr><td> 4                                      </td><td>000063d3fe1cf2ba248b9e3c3f0334845a27a6bf</td><td>M                                       </td><td>[18-24]                                 </td></tr>\n",
       "\t<tr><td> 5                                      </td><td>00007a47085b9aab8af55f52ec8846ac479ac4fe</td><td>M                                       </td><td>[25-34]                                 </td></tr>\n",
       "\t<tr><td> 6                                      </td><td>0000c176103e538d5c9828e695fed4f7ae42dd01</td><td>M                                       </td><td>[18-24]                                 </td></tr>\n",
       "\t<tr><td> 7                                      </td><td>0000ee7dd906373efa37f4e1185bfe1e3f8695ae</td><td>M                                       </td><td>[1-17]                                  </td></tr>\n",
       "\t<tr><td> 8                                      </td><td>0000ef373bbd0d89ce796abae961f2705e8c1faf</td><td>F                                       </td><td>[18-24]                                 </td></tr>\n",
       "\t<tr><td> 9                                      </td><td>0000f687d4fe9c1ed49620fbc5ed5b0d7798ea20</td><td>F                                       </td><td>[18-24]                                 </td></tr>\n",
       "\t<tr><td>10                                      </td><td>0001399387da41d557219578fb08b12afa25ab67</td><td>M                                       </td><td>NA                                      </td></tr>\n",
       "\t<tr><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td></tr>\n",
       "\t<tr><td>359338                                  </td><td>fffd3df081c03829a9ea4699e2d85e868d7a791b</td><td>F                                       </td><td>[1-17]                                  </td></tr>\n",
       "\t<tr><td>359339                                  </td><td>fffe1f0fc9eb1432b98eb25b2c950850796a4a8b</td><td>M                                       </td><td>[25-34]                                 </td></tr>\n",
       "\t<tr><td>359340                                  </td><td>fffe356b9dab2fae1f887fabb1f08ab4976c91bb</td><td>F                                       </td><td>[18-24]                                 </td></tr>\n",
       "\t<tr><td>359341                                  </td><td>fffe3e8eb1b2db8dcb3f3fe753552d5803d085f1</td><td>F                                       </td><td>[18-24]                                 </td></tr>\n",
       "\t<tr><td>359342                                  </td><td>fffe454af08a58c7bafe77491b586711f6b5e6f1</td><td>M                                       </td><td>[25-34]                                 </td></tr>\n",
       "\t<tr><td>359343                                  </td><td>fffe7823f67b433b45f22056467db921c1d3d7d0</td><td>M                                       </td><td>[25-34]                                 </td></tr>\n",
       "\t<tr><td>359344                                  </td><td>fffe8637bd8234309e871409c7ebef99a720afc1</td><td>M                                       </td><td>[25-34]                                 </td></tr>\n",
       "\t<tr><td>359345                                  </td><td>fffe8c7f952d9b960a56ed4dcb40a415d924b224</td><td>M                                       </td><td>[18-24]                                 </td></tr>\n",
       "\t<tr><td>359346                                  </td><td>ffff9af9ae04d263dae91cb838b1f3a6725f5ffb</td><td>M                                       </td><td>[18-24]                                 </td></tr>\n",
       "\t<tr><td>359347                                  </td><td>ffff9ef87a7d9494ada2f9ade4b9ff637c0759ac</td><td>M                                       </td><td>[18-24]                                 </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|llll}\n",
       " id & key & gender & age\\\\\n",
       "\\hline\n",
       "\t  1                                       & 00000c289a1829a808ac09c00daf10bc3c4e223b & F                                        & {[}18-24{]}                             \\\\\n",
       "\t  2                                       & 00001411dc427966b17297bf4d69e7e193135d89 & F                                        & NA                                      \\\\\n",
       "\t  3                                       & 00004d2ac9316e22dc007ab2243d6fcb239e707d & NA                                       & NA                                      \\\\\n",
       "\t  4                                       & 000063d3fe1cf2ba248b9e3c3f0334845a27a6bf & M                                        & {[}18-24{]}                             \\\\\n",
       "\t  5                                       & 00007a47085b9aab8af55f52ec8846ac479ac4fe & M                                        & {[}25-34{]}                             \\\\\n",
       "\t  6                                       & 0000c176103e538d5c9828e695fed4f7ae42dd01 & M                                        & {[}18-24{]}                             \\\\\n",
       "\t  7                                       & 0000ee7dd906373efa37f4e1185bfe1e3f8695ae & M                                        & {[}1-17{]}                              \\\\\n",
       "\t  8                                       & 0000ef373bbd0d89ce796abae961f2705e8c1faf & F                                        & {[}18-24{]}                             \\\\\n",
       "\t  9                                       & 0000f687d4fe9c1ed49620fbc5ed5b0d7798ea20 & F                                        & {[}18-24{]}                             \\\\\n",
       "\t 10                                       & 0001399387da41d557219578fb08b12afa25ab67 & M                                        & NA                                      \\\\\n",
       "\t ⋮ & ⋮ & ⋮ & ⋮\\\\\n",
       "\t 359338                                   & fffd3df081c03829a9ea4699e2d85e868d7a791b & F                                        & {[}1-17{]}                              \\\\\n",
       "\t 359339                                   & fffe1f0fc9eb1432b98eb25b2c950850796a4a8b & M                                        & {[}25-34{]}                             \\\\\n",
       "\t 359340                                   & fffe356b9dab2fae1f887fabb1f08ab4976c91bb & F                                        & {[}18-24{]}                             \\\\\n",
       "\t 359341                                   & fffe3e8eb1b2db8dcb3f3fe753552d5803d085f1 & F                                        & {[}18-24{]}                             \\\\\n",
       "\t 359342                                   & fffe454af08a58c7bafe77491b586711f6b5e6f1 & M                                        & {[}25-34{]}                             \\\\\n",
       "\t 359343                                   & fffe7823f67b433b45f22056467db921c1d3d7d0 & M                                        & {[}25-34{]}                             \\\\\n",
       "\t 359344                                   & fffe8637bd8234309e871409c7ebef99a720afc1 & M                                        & {[}25-34{]}                             \\\\\n",
       "\t 359345                                   & fffe8c7f952d9b960a56ed4dcb40a415d924b224 & M                                        & {[}18-24{]}                             \\\\\n",
       "\t 359346                                   & ffff9af9ae04d263dae91cb838b1f3a6725f5ffb & M                                        & {[}18-24{]}                             \\\\\n",
       "\t 359347                                   & ffff9ef87a7d9494ada2f9ade4b9ff637c0759ac & M                                        & {[}18-24{]}                             \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "id | key | gender | age | \n",
       "|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n",
       "|  1                                       | 00000c289a1829a808ac09c00daf10bc3c4e223b | F                                        | [18-24]                                  | \n",
       "|  2                                       | 00001411dc427966b17297bf4d69e7e193135d89 | F                                        | NA                                       | \n",
       "|  3                                       | 00004d2ac9316e22dc007ab2243d6fcb239e707d | NA                                       | NA                                       | \n",
       "|  4                                       | 000063d3fe1cf2ba248b9e3c3f0334845a27a6bf | M                                        | [18-24]                                  | \n",
       "|  5                                       | 00007a47085b9aab8af55f52ec8846ac479ac4fe | M                                        | [25-34]                                  | \n",
       "|  6                                       | 0000c176103e538d5c9828e695fed4f7ae42dd01 | M                                        | [18-24]                                  | \n",
       "|  7                                       | 0000ee7dd906373efa37f4e1185bfe1e3f8695ae | M                                        | [1-17]                                   | \n",
       "|  8                                       | 0000ef373bbd0d89ce796abae961f2705e8c1faf | F                                        | [18-24]                                  | \n",
       "|  9                                       | 0000f687d4fe9c1ed49620fbc5ed5b0d7798ea20 | F                                        | [18-24]                                  | \n",
       "| 10                                       | 0001399387da41d557219578fb08b12afa25ab67 | M                                        | NA                                       | \n",
       "| ⋮ | ⋮ | ⋮ | ⋮ | \n",
       "| 359338                                   | fffd3df081c03829a9ea4699e2d85e868d7a791b | F                                        | [1-17]                                   | \n",
       "| 359339                                   | fffe1f0fc9eb1432b98eb25b2c950850796a4a8b | M                                        | [25-34]                                  | \n",
       "| 359340                                   | fffe356b9dab2fae1f887fabb1f08ab4976c91bb | F                                        | [18-24]                                  | \n",
       "| 359341                                   | fffe3e8eb1b2db8dcb3f3fe753552d5803d085f1 | F                                        | [18-24]                                  | \n",
       "| 359342                                   | fffe454af08a58c7bafe77491b586711f6b5e6f1 | M                                        | [25-34]                                  | \n",
       "| 359343                                   | fffe7823f67b433b45f22056467db921c1d3d7d0 | M                                        | [25-34]                                  | \n",
       "| 359344                                   | fffe8637bd8234309e871409c7ebef99a720afc1 | M                                        | [25-34]                                  | \n",
       "| 359345                                   | fffe8c7f952d9b960a56ed4dcb40a415d924b224 | M                                        | [18-24]                                  | \n",
       "| 359346                                   | ffff9af9ae04d263dae91cb838b1f3a6725f5ffb | M                                        | [18-24]                                  | \n",
       "| 359347                                   | ffff9ef87a7d9494ada2f9ade4b9ff637c0759ac | M                                        | [18-24]                                  | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "       id     key                                      gender age    \n",
       "1       1     00000c289a1829a808ac09c00daf10bc3c4e223b F      [18-24]\n",
       "2       2     00001411dc427966b17297bf4d69e7e193135d89 F      NA     \n",
       "3       3     00004d2ac9316e22dc007ab2243d6fcb239e707d NA     NA     \n",
       "4       4     000063d3fe1cf2ba248b9e3c3f0334845a27a6bf M      [18-24]\n",
       "5       5     00007a47085b9aab8af55f52ec8846ac479ac4fe M      [25-34]\n",
       "6       6     0000c176103e538d5c9828e695fed4f7ae42dd01 M      [18-24]\n",
       "7       7     0000ee7dd906373efa37f4e1185bfe1e3f8695ae M      [1-17] \n",
       "8       8     0000ef373bbd0d89ce796abae961f2705e8c1faf F      [18-24]\n",
       "9       9     0000f687d4fe9c1ed49620fbc5ed5b0d7798ea20 F      [18-24]\n",
       "10     10     0001399387da41d557219578fb08b12afa25ab67 M      NA     \n",
       "⋮      ⋮      ⋮                                        ⋮      ⋮      \n",
       "359338 359338 fffd3df081c03829a9ea4699e2d85e868d7a791b F      [1-17] \n",
       "359339 359339 fffe1f0fc9eb1432b98eb25b2c950850796a4a8b M      [25-34]\n",
       "359340 359340 fffe356b9dab2fae1f887fabb1f08ab4976c91bb F      [18-24]\n",
       "359341 359341 fffe3e8eb1b2db8dcb3f3fe753552d5803d085f1 F      [18-24]\n",
       "359342 359342 fffe454af08a58c7bafe77491b586711f6b5e6f1 M      [25-34]\n",
       "359343 359343 fffe7823f67b433b45f22056467db921c1d3d7d0 M      [25-34]\n",
       "359344 359344 fffe8637bd8234309e871409c7ebef99a720afc1 M      [25-34]\n",
       "359345 359345 fffe8c7f952d9b960a56ed4dcb40a415d924b224 M      [18-24]\n",
       "359346 359346 ffff9af9ae04d263dae91cb838b1f3a6725f5ffb M      [18-24]\n",
       "359347 359347 ffff9ef87a7d9494ada2f9ade4b9ff637c0759ac M      [18-24]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "##### id,key,gender,age\n",
    "\n",
    "usersLastfm = read_delim(\"data/lastfm/lastfm-users.csv\", delim=\",\", \n",
    "                   col_names=c(\"id\", \"key\", \"gender\", \"age\") )\n",
    "#usersLastfm <-  subset(usersLastfm, select = c(\"user\", \"gender\", \"age\"))\n",
    "#usersLastfm$age <- as.integer(usersLastfm$age)\n",
    "usersLastfm$age <- as.numeric(as.character(usersLastfm$age))\n",
    "usersLastfm$gender[usersLastfm$gender == \"f\"] <- \"F\"\n",
    "usersLastfm$gender[usersLastfm$gender == \"m\"] <- \"M\"\n",
    "usersLastfm$age[ as.numeric(usersLastfm$age) < 18] <- \"[1-17]\"\n",
    "usersLastfm$age[as.numeric(usersLastfm$age) >= 18 & as.numeric(usersLastfm$age) < 25] <- \"[18-24]\"\n",
    "usersLastfm$age[as.numeric(usersLastfm$age) >= 25 &  as.numeric(usersLastfm$age) < 35] <- \"[25-34]\"\n",
    "usersLastfm$age[as.numeric(usersLastfm$age) >= 35 &  as.numeric(usersLastfm$age) < 45] <- \"[35-44]\"\n",
    "usersLastfm$age[as.numeric(usersLastfm$age) >= 45 & as.numeric(usersLastfm$age) < 56] <- \"[45-55]\"\n",
    "usersLastfm$age[as.numeric(usersLastfm$age) >= 56 ] <- \"[56+]\"\n",
    "#usersLastfm$dataset = as.factor(\"LastFM\")\n",
    "usersLastfm$age = as.factor(usersLastfm$age)\n",
    "\n",
    "users.meta=usersLastfm\n",
    "usersLastfm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Parsed with column specification:\n",
      "cols(\n",
      "  DataSet = col_character(),\n",
      "  Partition = col_integer(),\n",
      "  Algorithm = col_character(),\n",
      "  User = col_integer(),\n",
      "  TestTime = col_double(),\n",
      "  TopN.nDCG = col_double(),\n",
      "  Rank = col_integer(),\n",
      "  RecipRank = col_double(),\n",
      "  AvgPrec = col_double()\n",
      ")\n",
      "Joining, by = \"user\"\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>DataSet</th><th scope=col>Partition</th><th scope=col>Algorithm</th><th scope=col>user</th><th scope=col>TestTime</th><th scope=col>TopN.nDCG</th><th scope=col>Rank</th><th scope=col>RecipRank</th><th scope=col>AvgPrec</th><th scope=col>gender</th><th scope=col>age</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>LastFM.UI</td><td>2        </td><td>PopCt    </td><td>11       </td><td>0.274    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>M        </td><td>[25-34]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>2        </td><td>Pop      </td><td>11       </td><td>0.287    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>M        </td><td>[25-34]  </td></tr>\n",
       "\t<tr><td>LastFM   </td><td>1        </td><td>Pop      </td><td>69       </td><td>0.452    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>F        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>3        </td><td>Pop      </td><td> 1       </td><td>0.350    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>F        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>3        </td><td>PopCt    </td><td> 1       </td><td>0.341    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>F        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>4        </td><td>Pop      </td><td>70       </td><td>0.303    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>F        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>2        </td><td>PopCt    </td><td>21       </td><td>0.093    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>M        </td><td>NA       </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>4        </td><td>PopCt    </td><td>70       </td><td>0.283    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>F        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>2        </td><td>Pop      </td><td>21       </td><td>0.089    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>M        </td><td>NA       </td></tr>\n",
       "\t<tr><td>LastFM   </td><td>1        </td><td>Pop      </td><td>92       </td><td>0.086    </td><td>NaN      </td><td> 3       </td><td>0.3333333</td><td>0.1766667</td><td>M        </td><td>[25-34]  </td></tr>\n",
       "\t<tr><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td></tr>\n",
       "\t<tr><td>LastFM    </td><td>1         </td><td>II-B      </td><td>358474    </td><td>1.182     </td><td>NaN       </td><td>20        </td><td>0.05000000</td><td>0.04318182</td><td>M         </td><td>[18-24]   </td></tr>\n",
       "\t<tr><td>LastFM    </td><td>1         </td><td>II-B      </td><td>358502    </td><td>1.367     </td><td>NaN       </td><td> 3        </td><td>0.33333333</td><td>0.24982747</td><td>F         </td><td>[18-24]   </td></tr>\n",
       "\t<tr><td>LastFM    </td><td>1         </td><td>II-B      </td><td>358533    </td><td>1.495     </td><td>NaN       </td><td> 1        </td><td>1.00000000</td><td>0.36844505</td><td>M         </td><td>[25-34]   </td></tr>\n",
       "\t<tr><td>LastFM    </td><td>1         </td><td>II-B      </td><td>358581    </td><td>1.551     </td><td>NaN       </td><td>11        </td><td>0.09090909</td><td>0.09090909</td><td>M         </td><td>[25-34]   </td></tr>\n",
       "\t<tr><td>LastFM    </td><td>1         </td><td>II-B      </td><td>358587    </td><td>1.615     </td><td>NaN       </td><td> 1        </td><td>1.00000000</td><td>1.00000000</td><td>M         </td><td>NA        </td></tr>\n",
       "\t<tr><td>LastFM    </td><td>1         </td><td>II-B      </td><td>358646    </td><td>1.559     </td><td>NaN       </td><td> 4        </td><td>0.25000000</td><td>0.14500000</td><td>F         </td><td>[25-34]   </td></tr>\n",
       "\t<tr><td>LastFM    </td><td>1         </td><td>II-B      </td><td>358661    </td><td>1.494     </td><td>NaN       </td><td>20        </td><td>0.05000000</td><td>0.04329237</td><td>M         </td><td>[25-34]   </td></tr>\n",
       "\t<tr><td>LastFM    </td><td>1         </td><td>II-B      </td><td>358667    </td><td>1.588     </td><td>NaN       </td><td>17        </td><td>0.05882353</td><td>0.06181564</td><td>M         </td><td>[18-24]   </td></tr>\n",
       "\t<tr><td>LastFM    </td><td>1         </td><td>II-B      </td><td>358707    </td><td>1.479     </td><td>NaN       </td><td>37        </td><td>0.02702703</td><td>0.02702703</td><td>M         </td><td>[25-34]   </td></tr>\n",
       "\t<tr><td>LastFM    </td><td>1         </td><td>II-B      </td><td>358714    </td><td>1.461     </td><td>NaN       </td><td> 2        </td><td>0.50000000</td><td>0.27696742</td><td>M         </td><td>[18-24]   </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|lllllllllll}\n",
       " DataSet & Partition & Algorithm & user & TestTime & TopN.nDCG & Rank & RecipRank & AvgPrec & gender & age\\\\\n",
       "\\hline\n",
       "\t LastFM.UI   & 2           & PopCt       & 11          & 0.274       & NaN         & NA          & 0.0000000   & 0.0000000   & M           & {[}25-34{]}\\\\\n",
       "\t LastFM.UI   & 2           & Pop         & 11          & 0.287       & NaN         & NA          & 0.0000000   & 0.0000000   & M           & {[}25-34{]}\\\\\n",
       "\t LastFM      & 1           & Pop         & 69          & 0.452       & NaN         & NA          & 0.0000000   & 0.0000000   & F           & {[}18-24{]}\\\\\n",
       "\t LastFM.UI   & 3           & Pop         &  1          & 0.350       & NaN         & NA          & 0.0000000   & 0.0000000   & F           & {[}18-24{]}\\\\\n",
       "\t LastFM.UI   & 3           & PopCt       &  1          & 0.341       & NaN         & NA          & 0.0000000   & 0.0000000   & F           & {[}18-24{]}\\\\\n",
       "\t LastFM.UI   & 4           & Pop         & 70          & 0.303       & NaN         & NA          & 0.0000000   & 0.0000000   & F           & {[}18-24{]}\\\\\n",
       "\t LastFM.UI & 2         & PopCt     & 21        & 0.093     & NaN       & NA        & 0.0000000 & 0.0000000 & M         & NA       \\\\\n",
       "\t LastFM.UI   & 4           & PopCt       & 70          & 0.283       & NaN         & NA          & 0.0000000   & 0.0000000   & F           & {[}18-24{]}\\\\\n",
       "\t LastFM.UI & 2         & Pop       & 21        & 0.089     & NaN       & NA        & 0.0000000 & 0.0000000 & M         & NA       \\\\\n",
       "\t LastFM      & 1           & Pop         & 92          & 0.086       & NaN         &  3          & 0.3333333   & 0.1766667   & M           & {[}25-34{]}\\\\\n",
       "\t ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮\\\\\n",
       "\t LastFM      & 1           & II-B        & 358474      & 1.182       & NaN         & 20          & 0.05000000  & 0.04318182  & M           & {[}18-24{]}\\\\\n",
       "\t LastFM      & 1           & II-B        & 358502      & 1.367       & NaN         &  3          & 0.33333333  & 0.24982747  & F           & {[}18-24{]}\\\\\n",
       "\t LastFM      & 1           & II-B        & 358533      & 1.495       & NaN         &  1          & 1.00000000  & 0.36844505  & M           & {[}25-34{]}\\\\\n",
       "\t LastFM      & 1           & II-B        & 358581      & 1.551       & NaN         & 11          & 0.09090909  & 0.09090909  & M           & {[}25-34{]}\\\\\n",
       "\t LastFM     & 1          & II-B       & 358587     & 1.615      & NaN        &  1         & 1.00000000 & 1.00000000 & M          & NA        \\\\\n",
       "\t LastFM      & 1           & II-B        & 358646      & 1.559       & NaN         &  4          & 0.25000000  & 0.14500000  & F           & {[}25-34{]}\\\\\n",
       "\t LastFM      & 1           & II-B        & 358661      & 1.494       & NaN         & 20          & 0.05000000  & 0.04329237  & M           & {[}25-34{]}\\\\\n",
       "\t LastFM      & 1           & II-B        & 358667      & 1.588       & NaN         & 17          & 0.05882353  & 0.06181564  & M           & {[}18-24{]}\\\\\n",
       "\t LastFM      & 1           & II-B        & 358707      & 1.479       & NaN         & 37          & 0.02702703  & 0.02702703  & M           & {[}25-34{]}\\\\\n",
       "\t LastFM      & 1           & II-B        & 358714      & 1.461       & NaN         &  2          & 0.50000000  & 0.27696742  & M           & {[}18-24{]}\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "DataSet | Partition | Algorithm | user | TestTime | TopN.nDCG | Rank | RecipRank | AvgPrec | gender | age | \n",
       "|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n",
       "| LastFM.UI | 2         | PopCt     | 11        | 0.274     | NaN       | NA        | 0.0000000 | 0.0000000 | M         | [25-34]   | \n",
       "| LastFM.UI | 2         | Pop       | 11        | 0.287     | NaN       | NA        | 0.0000000 | 0.0000000 | M         | [25-34]   | \n",
       "| LastFM    | 1         | Pop       | 69        | 0.452     | NaN       | NA        | 0.0000000 | 0.0000000 | F         | [18-24]   | \n",
       "| LastFM.UI | 3         | Pop       |  1        | 0.350     | NaN       | NA        | 0.0000000 | 0.0000000 | F         | [18-24]   | \n",
       "| LastFM.UI | 3         | PopCt     |  1        | 0.341     | NaN       | NA        | 0.0000000 | 0.0000000 | F         | [18-24]   | \n",
       "| LastFM.UI | 4         | Pop       | 70        | 0.303     | NaN       | NA        | 0.0000000 | 0.0000000 | F         | [18-24]   | \n",
       "| LastFM.UI | 2         | PopCt     | 21        | 0.093     | NaN       | NA        | 0.0000000 | 0.0000000 | M         | NA        | \n",
       "| LastFM.UI | 4         | PopCt     | 70        | 0.283     | NaN       | NA        | 0.0000000 | 0.0000000 | F         | [18-24]   | \n",
       "| LastFM.UI | 2         | Pop       | 21        | 0.089     | NaN       | NA        | 0.0000000 | 0.0000000 | M         | NA        | \n",
       "| LastFM    | 1         | Pop       | 92        | 0.086     | NaN       |  3        | 0.3333333 | 0.1766667 | M         | [25-34]   | \n",
       "| ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | \n",
       "| LastFM     | 1          | II-B       | 358474     | 1.182      | NaN        | 20         | 0.05000000 | 0.04318182 | M          | [18-24]    | \n",
       "| LastFM     | 1          | II-B       | 358502     | 1.367      | NaN        |  3         | 0.33333333 | 0.24982747 | F          | [18-24]    | \n",
       "| LastFM     | 1          | II-B       | 358533     | 1.495      | NaN        |  1         | 1.00000000 | 0.36844505 | M          | [25-34]    | \n",
       "| LastFM     | 1          | II-B       | 358581     | 1.551      | NaN        | 11         | 0.09090909 | 0.09090909 | M          | [25-34]    | \n",
       "| LastFM     | 1          | II-B       | 358587     | 1.615      | NaN        |  1         | 1.00000000 | 1.00000000 | M          | NA         | \n",
       "| LastFM     | 1          | II-B       | 358646     | 1.559      | NaN        |  4         | 0.25000000 | 0.14500000 | F          | [25-34]    | \n",
       "| LastFM     | 1          | II-B       | 358661     | 1.494      | NaN        | 20         | 0.05000000 | 0.04329237 | M          | [25-34]    | \n",
       "| LastFM     | 1          | II-B       | 358667     | 1.588      | NaN        | 17         | 0.05882353 | 0.06181564 | M          | [18-24]    | \n",
       "| LastFM     | 1          | II-B       | 358707     | 1.479      | NaN        | 37         | 0.02702703 | 0.02702703 | M          | [25-34]    | \n",
       "| LastFM     | 1          | II-B       | 358714     | 1.461      | NaN        |  2         | 0.50000000 | 0.27696742 | M          | [18-24]    | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "       DataSet   Partition Algorithm user   TestTime TopN.nDCG Rank RecipRank \n",
       "1      LastFM.UI 2         PopCt     11     0.274    NaN       NA   0.0000000 \n",
       "2      LastFM.UI 2         Pop       11     0.287    NaN       NA   0.0000000 \n",
       "3      LastFM    1         Pop       69     0.452    NaN       NA   0.0000000 \n",
       "4      LastFM.UI 3         Pop        1     0.350    NaN       NA   0.0000000 \n",
       "5      LastFM.UI 3         PopCt      1     0.341    NaN       NA   0.0000000 \n",
       "6      LastFM.UI 4         Pop       70     0.303    NaN       NA   0.0000000 \n",
       "7      LastFM.UI 2         PopCt     21     0.093    NaN       NA   0.0000000 \n",
       "8      LastFM.UI 4         PopCt     70     0.283    NaN       NA   0.0000000 \n",
       "9      LastFM.UI 2         Pop       21     0.089    NaN       NA   0.0000000 \n",
       "10     LastFM    1         Pop       92     0.086    NaN        3   0.3333333 \n",
       "⋮      ⋮         ⋮         ⋮         ⋮      ⋮        ⋮         ⋮    ⋮         \n",
       "296415 LastFM    1         II-B      358474 1.182    NaN       20   0.05000000\n",
       "296416 LastFM    1         II-B      358502 1.367    NaN        3   0.33333333\n",
       "296417 LastFM    1         II-B      358533 1.495    NaN        1   1.00000000\n",
       "296418 LastFM    1         II-B      358581 1.551    NaN       11   0.09090909\n",
       "296419 LastFM    1         II-B      358587 1.615    NaN        1   1.00000000\n",
       "296420 LastFM    1         II-B      358646 1.559    NaN        4   0.25000000\n",
       "296421 LastFM    1         II-B      358661 1.494    NaN       20   0.05000000\n",
       "296422 LastFM    1         II-B      358667 1.588    NaN       17   0.05882353\n",
       "296423 LastFM    1         II-B      358707 1.479    NaN       37   0.02702703\n",
       "296424 LastFM    1         II-B      358714 1.461    NaN        2   0.50000000\n",
       "       AvgPrec    gender age    \n",
       "1      0.0000000  M      [25-34]\n",
       "2      0.0000000  M      [25-34]\n",
       "3      0.0000000  F      [18-24]\n",
       "4      0.0000000  F      [18-24]\n",
       "5      0.0000000  F      [18-24]\n",
       "6      0.0000000  F      [18-24]\n",
       "7      0.0000000  M      NA     \n",
       "8      0.0000000  F      [18-24]\n",
       "9      0.0000000  M      NA     \n",
       "10     0.1766667  M      [25-34]\n",
       "⋮      ⋮          ⋮      ⋮      \n",
       "296415 0.04318182 M      [18-24]\n",
       "296416 0.24982747 F      [18-24]\n",
       "296417 0.36844505 M      [25-34]\n",
       "296418 0.09090909 M      [25-34]\n",
       "296419 1.00000000 M      NA     \n",
       "296420 0.14500000 F      [25-34]\n",
       "296421 0.04329237 M      [25-34]\n",
       "296422 0.06181564 M      [18-24]\n",
       "296423 0.02702703 M      [25-34]\n",
       "296424 0.27696742 M      [18-24]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "user.results.lastFm = read_csv(\"data/lastfm/lastfm-user-results.csv\", guess_max=10000) %>%\n",
    "    rename(user=User) %>%\n",
    "    inner_join(users.meta %>% rename(user=id) %>% select(-key))\n",
    "user.results.lastFm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Concat Datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>DataSet</th><th scope=col>Partition</th><th scope=col>Algorithm</th><th scope=col>user</th><th scope=col>TestTime</th><th scope=col>TopN.nDCG</th><th scope=col>Rank</th><th scope=col>RecipRank</th><th scope=col>AvgPrec</th><th scope=col>gender</th><th scope=col>age</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>LastFM.UI</td><td>2        </td><td>PopCt    </td><td>11       </td><td>0.274    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>M        </td><td>[25-34]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>2        </td><td>Pop      </td><td>11       </td><td>0.287    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>M        </td><td>[25-34]  </td></tr>\n",
       "\t<tr><td>LastFM   </td><td>1        </td><td>Pop      </td><td>69       </td><td>0.452    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>F        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>3        </td><td>Pop      </td><td> 1       </td><td>0.350    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>F        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>3        </td><td>PopCt    </td><td> 1       </td><td>0.341    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>F        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>4        </td><td>Pop      </td><td>70       </td><td>0.303    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>F        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>2        </td><td>PopCt    </td><td>21       </td><td>0.093    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>M        </td><td>NA       </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>4        </td><td>PopCt    </td><td>70       </td><td>0.283    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>F        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>LastFM.UI</td><td>2        </td><td>Pop      </td><td>21       </td><td>0.089    </td><td>NaN      </td><td>NA       </td><td>0.0000000</td><td>0.0000000</td><td>M        </td><td>NA       </td></tr>\n",
       "\t<tr><td>LastFM   </td><td>1        </td><td>Pop      </td><td>92       </td><td>0.086    </td><td>NaN      </td><td> 3       </td><td>0.3333333</td><td>0.1766667</td><td>M        </td><td>[25-34]  </td></tr>\n",
       "\t<tr><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td><td>⋮</td></tr>\n",
       "\t<tr><td>Lenskit1M</td><td>1        </td><td>UU       </td><td>5999     </td><td>0.690    </td><td>0        </td><td>NA       </td><td>0        </td><td>0        </td><td>F        </td><td>NA       </td></tr>\n",
       "\t<tr><td>Lenskit1M</td><td>1        </td><td>UU       </td><td>6007     </td><td>0.787    </td><td>0        </td><td>NA       </td><td>0        </td><td>0        </td><td>M        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>Lenskit1M</td><td>1        </td><td>UU       </td><td>6008     </td><td>0.859    </td><td>0        </td><td>NA       </td><td>0        </td><td>0        </td><td>NA       </td><td>[25-34]  </td></tr>\n",
       "\t<tr><td>Lenskit1M</td><td>1        </td><td>UU       </td><td>6012     </td><td>0.681    </td><td>0        </td><td>NA       </td><td>0        </td><td>0        </td><td>M        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>Lenskit1M</td><td>1        </td><td>UU       </td><td>6013     </td><td>0.820    </td><td>0        </td><td>NA       </td><td>0        </td><td>0        </td><td>M        </td><td>[25-34]  </td></tr>\n",
       "\t<tr><td>Lenskit1M</td><td>1        </td><td>UU       </td><td>6018     </td><td>0.907    </td><td>0        </td><td>NA       </td><td>0        </td><td>0        </td><td>NA       </td><td>NA       </td></tr>\n",
       "\t<tr><td>Lenskit1M</td><td>1        </td><td>UU       </td><td>6022     </td><td>0.856    </td><td>0        </td><td>NA       </td><td>0        </td><td>0        </td><td>F        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>Lenskit1M</td><td>1        </td><td>UU       </td><td>6032     </td><td>0.788    </td><td>0        </td><td>NA       </td><td>0        </td><td>0        </td><td>M        </td><td>[18-24]  </td></tr>\n",
       "\t<tr><td>Lenskit1M</td><td>1        </td><td>UU       </td><td>6034     </td><td>0.762    </td><td>0        </td><td>NA       </td><td>0        </td><td>0        </td><td>M        </td><td>NA       </td></tr>\n",
       "\t<tr><td>Lenskit1M</td><td>1        </td><td>UU       </td><td>6035     </td><td>0.868    </td><td>0        </td><td>NA       </td><td>0        </td><td>0        </td><td>M        </td><td>[25-34]  </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|lllllllllll}\n",
       " DataSet & Partition & Algorithm & user & TestTime & TopN.nDCG & Rank & RecipRank & AvgPrec & gender & age\\\\\n",
       "\\hline\n",
       "\t LastFM.UI   & 2           & PopCt       & 11          & 0.274       & NaN         & NA          & 0.0000000   & 0.0000000   & M           & {[}25-34{]}\\\\\n",
       "\t LastFM.UI   & 2           & Pop         & 11          & 0.287       & NaN         & NA          & 0.0000000   & 0.0000000   & M           & {[}25-34{]}\\\\\n",
       "\t LastFM      & 1           & Pop         & 69          & 0.452       & NaN         & NA          & 0.0000000   & 0.0000000   & F           & {[}18-24{]}\\\\\n",
       "\t LastFM.UI   & 3           & Pop         &  1          & 0.350       & NaN         & NA          & 0.0000000   & 0.0000000   & F           & {[}18-24{]}\\\\\n",
       "\t LastFM.UI   & 3           & PopCt       &  1          & 0.341       & NaN         & NA          & 0.0000000   & 0.0000000   & F           & {[}18-24{]}\\\\\n",
       "\t LastFM.UI   & 4           & Pop         & 70          & 0.303       & NaN         & NA          & 0.0000000   & 0.0000000   & F           & {[}18-24{]}\\\\\n",
       "\t LastFM.UI & 2         & PopCt     & 21        & 0.093     & NaN       & NA        & 0.0000000 & 0.0000000 & M         & NA       \\\\\n",
       "\t LastFM.UI   & 4           & PopCt       & 70          & 0.283       & NaN         & NA          & 0.0000000   & 0.0000000   & F           & {[}18-24{]}\\\\\n",
       "\t LastFM.UI & 2         & Pop       & 21        & 0.089     & NaN       & NA        & 0.0000000 & 0.0000000 & M         & NA       \\\\\n",
       "\t LastFM      & 1           & Pop         & 92          & 0.086       & NaN         &  3          & 0.3333333   & 0.1766667   & M           & {[}25-34{]}\\\\\n",
       "\t ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮ & ⋮\\\\\n",
       "\t Lenskit1M & 1         & UU        & 5999      & 0.690     & 0         & NA        & 0         & 0         & F         & NA       \\\\\n",
       "\t Lenskit1M   & 1           & UU          & 6007        & 0.787       & 0           & NA          & 0           & 0           & M           & {[}18-24{]}\\\\\n",
       "\t Lenskit1M   & 1           & UU          & 6008        & 0.859       & 0           & NA          & 0           & 0           & NA          & {[}25-34{]}\\\\\n",
       "\t Lenskit1M   & 1           & UU          & 6012        & 0.681       & 0           & NA          & 0           & 0           & M           & {[}18-24{]}\\\\\n",
       "\t Lenskit1M   & 1           & UU          & 6013        & 0.820       & 0           & NA          & 0           & 0           & M           & {[}25-34{]}\\\\\n",
       "\t Lenskit1M & 1         & UU        & 6018      & 0.907     & 0         & NA        & 0         & 0         & NA        & NA       \\\\\n",
       "\t Lenskit1M   & 1           & UU          & 6022        & 0.856       & 0           & NA          & 0           & 0           & F           & {[}18-24{]}\\\\\n",
       "\t Lenskit1M   & 1           & UU          & 6032        & 0.788       & 0           & NA          & 0           & 0           & M           & {[}18-24{]}\\\\\n",
       "\t Lenskit1M & 1         & UU        & 6034      & 0.762     & 0         & NA        & 0         & 0         & M         & NA       \\\\\n",
       "\t Lenskit1M   & 1           & UU          & 6035        & 0.868       & 0           & NA          & 0           & 0           & M           & {[}25-34{]}\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "DataSet | Partition | Algorithm | user | TestTime | TopN.nDCG | Rank | RecipRank | AvgPrec | gender | age | \n",
       "|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|\n",
       "| LastFM.UI | 2         | PopCt     | 11        | 0.274     | NaN       | NA        | 0.0000000 | 0.0000000 | M         | [25-34]   | \n",
       "| LastFM.UI | 2         | Pop       | 11        | 0.287     | NaN       | NA        | 0.0000000 | 0.0000000 | M         | [25-34]   | \n",
       "| LastFM    | 1         | Pop       | 69        | 0.452     | NaN       | NA        | 0.0000000 | 0.0000000 | F         | [18-24]   | \n",
       "| LastFM.UI | 3         | Pop       |  1        | 0.350     | NaN       | NA        | 0.0000000 | 0.0000000 | F         | [18-24]   | \n",
       "| LastFM.UI | 3         | PopCt     |  1        | 0.341     | NaN       | NA        | 0.0000000 | 0.0000000 | F         | [18-24]   | \n",
       "| LastFM.UI | 4         | Pop       | 70        | 0.303     | NaN       | NA        | 0.0000000 | 0.0000000 | F         | [18-24]   | \n",
       "| LastFM.UI | 2         | PopCt     | 21        | 0.093     | NaN       | NA        | 0.0000000 | 0.0000000 | M         | NA        | \n",
       "| LastFM.UI | 4         | PopCt     | 70        | 0.283     | NaN       | NA        | 0.0000000 | 0.0000000 | F         | [18-24]   | \n",
       "| LastFM.UI | 2         | Pop       | 21        | 0.089     | NaN       | NA        | 0.0000000 | 0.0000000 | M         | NA        | \n",
       "| LastFM    | 1         | Pop       | 92        | 0.086     | NaN       |  3        | 0.3333333 | 0.1766667 | M         | [25-34]   | \n",
       "| ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | ⋮ | \n",
       "| Lenskit1M | 1         | UU        | 5999      | 0.690     | 0         | NA        | 0         | 0         | F         | NA        | \n",
       "| Lenskit1M | 1         | UU        | 6007      | 0.787     | 0         | NA        | 0         | 0         | M         | [18-24]   | \n",
       "| Lenskit1M | 1         | UU        | 6008      | 0.859     | 0         | NA        | 0         | 0         | NA        | [25-34]   | \n",
       "| Lenskit1M | 1         | UU        | 6012      | 0.681     | 0         | NA        | 0         | 0         | M         | [18-24]   | \n",
       "| Lenskit1M | 1         | UU        | 6013      | 0.820     | 0         | NA        | 0         | 0         | M         | [25-34]   | \n",
       "| Lenskit1M | 1         | UU        | 6018      | 0.907     | 0         | NA        | 0         | 0         | NA        | NA        | \n",
       "| Lenskit1M | 1         | UU        | 6022      | 0.856     | 0         | NA        | 0         | 0         | F         | [18-24]   | \n",
       "| Lenskit1M | 1         | UU        | 6032      | 0.788     | 0         | NA        | 0         | 0         | M         | [18-24]   | \n",
       "| Lenskit1M | 1         | UU        | 6034      | 0.762     | 0         | NA        | 0         | 0         | M         | NA        | \n",
       "| Lenskit1M | 1         | UU        | 6035      | 0.868     | 0         | NA        | 0         | 0         | M         | [25-34]   | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "       DataSet   Partition Algorithm user TestTime TopN.nDCG Rank RecipRank\n",
       "1      LastFM.UI 2         PopCt     11   0.274    NaN       NA   0.0000000\n",
       "2      LastFM.UI 2         Pop       11   0.287    NaN       NA   0.0000000\n",
       "3      LastFM    1         Pop       69   0.452    NaN       NA   0.0000000\n",
       "4      LastFM.UI 3         Pop        1   0.350    NaN       NA   0.0000000\n",
       "5      LastFM.UI 3         PopCt      1   0.341    NaN       NA   0.0000000\n",
       "6      LastFM.UI 4         Pop       70   0.303    NaN       NA   0.0000000\n",
       "7      LastFM.UI 2         PopCt     21   0.093    NaN       NA   0.0000000\n",
       "8      LastFM.UI 4         PopCt     70   0.283    NaN       NA   0.0000000\n",
       "9      LastFM.UI 2         Pop       21   0.089    NaN       NA   0.0000000\n",
       "10     LastFM    1         Pop       92   0.086    NaN        3   0.3333333\n",
       "⋮      ⋮         ⋮         ⋮         ⋮    ⋮        ⋮         ⋮    ⋮        \n",
       "362855 Lenskit1M 1         UU        5999 0.690    0         NA   0        \n",
       "362856 Lenskit1M 1         UU        6007 0.787    0         NA   0        \n",
       "362857 Lenskit1M 1         UU        6008 0.859    0         NA   0        \n",
       "362858 Lenskit1M 1         UU        6012 0.681    0         NA   0        \n",
       "362859 Lenskit1M 1         UU        6013 0.820    0         NA   0        \n",
       "362860 Lenskit1M 1         UU        6018 0.907    0         NA   0        \n",
       "362861 Lenskit1M 1         UU        6022 0.856    0         NA   0        \n",
       "362862 Lenskit1M 1         UU        6032 0.788    0         NA   0        \n",
       "362863 Lenskit1M 1         UU        6034 0.762    0         NA   0        \n",
       "362864 Lenskit1M 1         UU        6035 0.868    0         NA   0        \n",
       "       AvgPrec   gender age    \n",
       "1      0.0000000 M      [25-34]\n",
       "2      0.0000000 M      [25-34]\n",
       "3      0.0000000 F      [18-24]\n",
       "4      0.0000000 F      [18-24]\n",
       "5      0.0000000 F      [18-24]\n",
       "6      0.0000000 F      [18-24]\n",
       "7      0.0000000 M      NA     \n",
       "8      0.0000000 F      [18-24]\n",
       "9      0.0000000 M      NA     \n",
       "10     0.1766667 M      [25-34]\n",
       "⋮      ⋮         ⋮      ⋮      \n",
       "362855 0         F      NA     \n",
       "362856 0         M      [18-24]\n",
       "362857 0         NA     [25-34]\n",
       "362858 0         M      [18-24]\n",
       "362859 0         M      [25-34]\n",
       "362860 0         NA     NA     \n",
       "362861 0         F      [18-24]\n",
       "362862 0         M      [18-24]\n",
       "362863 0         M      NA     \n",
       "362864 0         M      [25-34]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "user.results <- rbind.fill(user.results.lastFm,user.results.lenskit1m)\n",
    "user.results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analyze by Gender"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using DataSet, Algorithm, gender as id variables\n"
     ]
    }
   ],
   "source": [
    "gender.results = user.results %>%\n",
    "    group_by(DataSet, Algorithm, gender) %>%\n",
    "    summarize( MRR=mean(RecipRank, na.rm=TRUE))\n",
    "gender.tall = melt(gender.results, variable.name = \"Metric\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 221,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using DataSet, Algorithm, gender as id variables\n"
     ]
    }
   ],
   "source": [
    "overall.results = user.results %>%\n",
    "    group_by(DataSet, Algorithm) %>%\n",
    "    summarize( MRR=mean(RecipRank, na.rm=TRUE))\n",
    "combined.results = rbind(gender.results, mutate(overall.results))\n",
    "combined.tall = melt(combined.results, variable.name='Metric')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "combined.tall$Algorithm[combined.tall$Algorithm == \"Pop\"] <- \"POP-B\"\n",
    "combined.tall$Algorithm[combined.tall$Algorithm == \"PopCt\"] <- \"PoP-C\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {},
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0gAAAJYCAIAAADXJFGjAAAACXBIWXMAABJ0AAASdAHeZh94\nAAAgAElEQVR4nO3deWBTVd7/8ZOtTZMudEFKEaiyWAGllILYlm0oFJBVEBGhlEXhAXkcQAR/\nPgKDOsAgUBUcZXMY64yUovRBwTKLCpWySlkeKDLsIAxlKXGaLmmS3x93jJkCtYXcpjm8X3/d\ne3LuybfxUD89d4nG6XQKAAAA+D6ttwsAAACAZxDsAAAAJEGwAwAAkATBDgAAQBIEOwAAAEkQ\n7AAAACRBsAMAAJAEwQ4AAEASem8XcCdKSkpKS0u9XYWc/P39jUZjcXFxRUWFt2uBDwsICPDz\n8/vxxx8dDoe3a4EPM5vNer3eYrHwLH2VWCwWlf6RPvDAA2oMi1/kk8HO6XTa7XZvVyEtrVbL\nJ4y7p9VqHQ4HEwl3Q6PRaLVau91OsFNJRUUF/0glw6lYAAAASRDsAAAAJEGwAwAAkATBDgAA\nQBIEOwAAAEkQ7AAAACRBsAMAAJAEwQ4AAEASBDsAAABJEOwAAAAkQbADAACQBMEOAABAEvpa\neA+Hw/HnP//573//u91uT0pKGjNmjE6nq9Tn6tWrK1asOHz4sFarjY+PHzt2bFBQUC3UBgAA\nII3aWLHLzMz84osvxo0bN2nSpO3bt69du7ZSB6fTuWjRosuXL0+fPn3q1KlHjx5dvnx5LRQG\nAAAgE9VX7CoqKjZv3jxq1KiEhAQhRFlZ2fLly0eMGGE0Gl19Ll26dOTIkfT09AcffFAIMXLk\nyMWLF9vt9psX9gAAAHA7qq/YnT9/vqioKC4uTtmNi4uzWq0nTpxw71NSUhIfH3///fcruyaT\nyel02mw2tWsDAACQieordteuXRNChIeHK7tms9loNBYVFbn3efDBB2fPni2EcDgchYWFn3/+\neWxsrPuSnsVimTx5smt30KBBAwcOVLvye5NWqxVCmM1mk8nk7Vrgw5SJFBwc7HQ6vV0LfJhy\n3iYkJMTbhUjrxo0b3i4BHqZ6sLNYLAaDQa//+Y1MJtPtZtJrr7126NChoKCgxYsXu7fb7faj\nR4+6djt37uw+IDyOk+DwCCYSPIJf+ED1qf6vJTAw0GazuV8wZ7VaAwMDb9l56tSp165d++KL\nL1566aUVK1a4Fo1CQ0P37t3r6ma1Wq9cuaJ25fcmk8lkMpksFkt5ebm3a4EPCwwMNBqN169f\nt9vt3q4FPiwkJMRgMFy9epWlX6CaVL/GLjQ0VPx0QlYIUVJSUlpaqjS6XL58+eTJk0KIiIiI\nli1b/vd//3dpaemhQ4fUrg0AAEAmqge76OjokJCQ/Px8ZTc/Pz8gIKBFixbufQoKCubMmeP6\ny15Z4dNoNGrXBgAAIBPVT8XqdLq+fftmZGRERkZqtdrVq1f36tVLuTEiJyenrKxswIABbdu2\nLS0tfffdd/v06VNRUbFhw4awsLA2bdqoXRsAAIBMauOK1OHDh9tstvT0dIfDkZSUlJaWprTn\n5eVZLJYBAwaEhITMnTs3MzNz3rx5Wq22VatWr7/+OndlApXs/by+GsPG9ytUY1gAQO3T+OIV\nqVar1Wq1ersKOXHzRF3mQ8GOmyfgEdw8obYbN26o9I+0ZcuWagyLX1QbXykGAACAWsDDgQDP\nC1o0T5VxH1ZlVACANFixAwAAkATBDgAAQBIEOwAAAEkQ7AAAACRBsAMAAJAEwQ4AAEASBDsA\nAABJ8Bw7ALi3+NBXmACoKVbsAAAAJMGKHQDURWp9f4ngK0wAmbFiBwAAIAmCHQAAgCQIdgAA\nAJIg2AEAAEiCYAcAACAJgh0AAIAkCHYAAACSINgBAABIgmAHAAAgCYIdAACAJAh2AAAAkiDY\nAQAASIJgBwAAIAmCHQAAgCQIdgAAAJLQe7uAe8Lez+urMWx8v0I1hgUAAD6KYPcfghbNU2Xc\nh1UZFQAAwB2nYgEAACRBsAMAAJAEwQ4AAEASBDsAAABJEOwAAAAkQbADAACQBMEOAABAEgQ7\nAAAASRDsAAAAJEGwAwAAkATBDgAAQBIEOwAAAEkQ7AAAACRBsAMAAJAEwQ4AAEASBDsAAABJ\nEOwAAAAkQbADAACQBMEOAABAEgQ7AAAASRDsAAAAJEGwAwAAkATBDgAAQBIEOwAAAEkQ7AAA\nACRBsAMAAJAEwQ4AAEASBDsAAABJEOwAAAAkQbADAACQBMEOAABAEgQ7AAAASRDsAAAAJEGw\nAwAAkATBDgAAQBIEOwAAAEkQ7AAAACRBsAMAAJAEwQ4AAEASBDsAAABJEOwAAAAkQbADAACQ\nBMEOAABAEgQ7AAAASRDsAAAAJKH3dgF3QqfTBQYGersK71PjQ9Dr9UIIo9Ho5+fn8cFRN6kx\nkQwGgxDCZDI5nU6PD466SY2JpNPphBBms9njI0Nx48YNb5cAD/PJYOdwOMrLy9UY2ajGoKop\nKytTY1i9Xm+z2SoqKtQY/B7BRNJqtTqdzmaz2e12jw9+j/CtWSTUmUh6vV6r1ZaXl/MXAlBN\nPhnsnE6nzWZTY2Tf+k2qxoegLLTY7XaVPuF7BBPJ399fGZlgd8d8axYJdSaSkudsNhvBDqgm\nrrEDAACQBMEOAABAEgQ7AAAASRDsAAAAJEGwAwAAkATBDgAAQBIEOwAAAEkQ7AAAACRBsAMA\nAJAEwQ4AAEASBDsAAABJEOwAAAAkQbADAACQBMEOAABAEgQ7AAAASRDsAAAAJEGwAwAAkATB\nDgAAQBIEOwAAAEkQ7AAAACRBsAMAAJAEwQ4AAEASBDsAAABJEOwAAAAkQbADAACQBMEOAABA\nEgQ7AAAASRDsAAAAJEGwAwAAkATBDgAAQBIEOwAAAEkQ7AAAACRBsAMAAJAEwQ4AAEASBDsA\nAABJEOwAAAAkQbADAACQBMEOAABAEgQ7AAAASRDsAAAAJEGwAwAAkATBDgAAQBIEOwAAAEkQ\n7AAAACRBsAMAAJAEwQ4AAEASBDsAAABJEOwAAAAkQbADAACQBMEOAABAEgQ7AAAASRDsAAAA\nJEGwAwAAkATBDgAAQBIEOwAAAEkQ7AAAACRBsAMAAJAEwQ4AAEASBDsAAABJEOwAAAAkQbAD\nAACQBMEOAABAEgQ7AAAASRDsAAAAJEGwAwAAkATBDgAAQBIEOwAAAEkQ7AAAgI+5cOGCVqvV\naDSLFy+u9FK3bt06deqkxpvGx8d369btli/17t07Pj5ejTetKYIdAADwMevXr3c6ncpGrb1p\nQEBAQECAsp2enq7RaK5cuVJr715Nem8XAAAAUDPr1q0zGo0dO3bctm3bmTNnmjZtWgtvun37\n9lp4l7vEih0AAPAlZ86c2blzZ0pKyjPPPCNqd9Gu7iPYAQAAX5KZmSmEGDJkSL9+/cQvBbuc\nnJxu3bqFhIQ8/vjjmZmZb731ltFodL26b9++vn37NmjQIDIysk+fPnv37nW9lJycPHTo0OPH\nj/fu3Ts6OloIkZSUpFxj161bt6lTpwoh6tevP3z4cNchJ0+e7N+/f/369SMjI8ePH3/jxg2l\nvXfv3oMGDdq7d2/Pnj3r1asXFxe3YcOG8vLyadOmNW/ePCgoqE+fPmfOnPHIh0OwAwAAvmTd\nunV6vb5fv373339/u3btdu/effr06Vv2zMzM7Nu377Vr16ZOnRobG5uWlpaVleV69a9//evj\njz9++PDhMWPGpKWlHT58+PHHH9+6daurQ1FRUf/+/c+fP9+7d2/3YdPT0ydPniyE2Lhx4+zZ\ns5XGH374oUuXLk2bNp0zZ06bNm1Wr1790ksvuQ45evTok08+2blz53nz5hUWFg4fPjwxMXHv\n3r0vvfTSs88+++WXXyoD3j2usQMAAD7jxIkT+/btS0lJCQ0NFUIMGDBg//7969evnzFjRqWe\nZWVlL730Utu2bXNzc00mkxBi4MCBffr08ff3F0I4HI6pU6dGRETs3bv3vvvuE0JMmzatbdu2\nL730Un5+vlarFUL87W9/mzBhwu9//3uNRuM+cmxsbPPmzYUQiYmJERERSuPFixfff//9CRMm\nCCEmT5782GOP/e1vf3Md8v3333/11VfKgl/Tpk0HDRpkt9v//ve/6/V6IcSxY8d27Njhkc+n\nNlbsHA7Hxx9/PG7cuLS0tFWrVtnt9pv7lJeXf/DBBxMmTBg+fPjcuXMvXLhQC4UBAADfsm7d\nOiHEkCFDlN3+/fuLn07OVrJz585z585NmzZNSXVCiJSUlDZt2ijbp0+fPnz48H/9138pqU4I\ncd99902cOPHQoUOus6LK41QqpbrbMZvNY8eOdR3Ypk0bq9XqerVhw4auR6U8/PDDQojhw4cr\nqU5pce98N2oj2GVmZn7xxRfjxo2bNGnS9u3b165de3OfxYsX79y5MzU19bXXXnM6na+99pqn\nfkIAACCNTz75RAjRtm3b06dPnz59OiwsLDw8fO/evadOnarU8/jx40IIV5ITQmg0mlatWinb\n//jHP4QQjzzyiPshyu6JEyeU3ejoaLPZXM3CoqOjDQaDa1dZ83MJCwur9NLNLR6herCrqKjY\nvHnzqFGjEhISOnbsOH78+K1bt5aWlrr3uXbtWl5e3qRJkxITE1u3bj1r1qwff/xxz549atcG\nAAB8yNGjRw8dOiSEeOyxxx544IEHHnjgwQcfvHr1qrjVLRQ2m+3mEVwRSnkM3i1fraioUHbd\ns9cvcr8nw4tUD3bnz58vKiqKi4tTduPi4qxWqysLKywWS/PmzVu2bKnsGo1Gf3//a9euqV0b\nAADwIcp52IkTJ37mZsmSJeJWZ2OVXHHkyBH3xqNHjyobzZo1E0IcPnzY/VUlNbZo0UKtH0B9\nqt88oeSz8PBwZddsNhuNxqKiIvc+0dHRyn8VxbfffmuxWJQz0AqbzbZ//37XbkREhGvAe5n7\nkq+nKH+s6HQ6NQZH3aTeRDIYDB48v4A6To2JpFzbZDAYbrm4gnuN0+lct26dTqebM2dOZGSk\nq93hcLz77rv79u07efLkgw8+6Grv0KFDRETE4sWLBw8erHxjxNatWw8cOKDcPPHAAw+0atXq\nvffemzBhQv369YUQly9ffu+991q1aqU83KQ6HA6HB39Aj1A92FksFoPB4Lo8UAhhMplcT3ap\nxG63b9q06Q9/+ENKSkpMTIyr/V//+tekSZNcu88///zzzz+vRrVlagyqmpCQEJVGrv4lBbgl\nJpIiMDBQpZHvBb41i4SaEyk4OFilkVEHvxGrCocOHSooKOjfv797qhNCaLXa1NTU3/zmN+vX\nr585c6arPTg4eMGCBePHj+/UqdOTTz55+fLl9evXx8fHFxQUCCF0Ot2SJUv69esXHx8/YsQI\np9P58ccfX7ly5Y9//KNOp/vFYpS/ZBYtWtS3b9/u3bt7+me9c6oHu8DAQJvNZrfbXR+T1Wq9\n5a/7M2fOLF68+NKlS+PGjVMeOegSEBAwZcoU127r1q2Li4vVqNa3nv6ixodgMBj8/PxKS0tv\nefMyqomJ5O/vr9frS0pK6uCfs77Ct2aRUGciGY1GnU6n0i98+BzlPOyYMWNufkkJdpmZme7B\nTggxbty4iIiI+fPnL1myJC4ubuPGjcuWLXNd65WSkvLtt9++9tpra9asEULExcV99tln8fHx\n1Smmf//+WVlZy5cvt1gsdSrYadRe3z5x4sTUqVNXr16trHOWlJQ8/fTTb775ZqX7UA4dOjR3\n7tx27dpNnjxZeTJNFaxWq0r3zAYtmqfGsF89/K4aw8b3K/T4mCaTyWQyWSyW8vJyjw9+72Ai\nBQYGGo3G69ev8xfCHVNpFgmfmkghISEGg+Hq1aucilXJjRs3VPpH6rpu3ovKysoOHTrUuHHj\nBg0auBq7d+9eWlqal5fnxcJUpfrlL9HR0SEhIfn5+cpufn5+QEBApcsSbTbbokWLevXq9eqr\nr/5iqgMAAPhFGo2ma9eu7mt4Z8+ezc3N/dWvfuXFqtSm+mK/Tqfr27dvRkZGZGSkVqtdvXp1\nr169lFuCc3JyysrKBgwYcODAgaKiohYtWrh/R1uTJk3cIzYAAED1+fn5TZkyZeHChXq9vkeP\nHufPn3/77bfNZrP7xV3yqY2rOIYPH26z2dLT0x0OR1JSUlpamtKel5dnsVgGDBigfM9Eenq6\n+1ETJkx44oknaqE8AAAgpTfeeKNBgwarVq3605/+FBUV1bVr1wULFlS690IytRHsNBpNampq\nampqpfa5c+cqGwMHDhw4cGAtVAIAAO4der1+6tSpU6dO9XYhtYdHTAEAAEiCYAcAACAJn3tS\nEgAAuEd5/OEsWq1W+YITaRDsAACAbygrK9P9ZpYHB9T/dml1vmfCh3AqFgAAQBIEOwAAAEkQ\n7AAAACRBsAMAAJAEwQ4AAEASBDsAAABJEOwAAAAkQbADAAD4t1mzZsXExNy8XUlMTIzmJwaD\nISYmZu3atbVY5m3xgGIAAIAaGzly5OTJk4UQVqs1Ozs7LS2tUaNGycnJ3q2KYAcAAFBjjRo1\n6tSpk7LdvXv3r7/+Oisry+vBjlOxAAAAd0Wj0QQFBYWGhnq7EIIdAABAzTkcjoqKioqKCovF\n8tFHHx06dGjYsGHeLopgBwAAUHOLFi0yGAwGgyEkJCQ1NXXSpEmxsbHeLopgBwAAUKXs7Ox6\nPzly5IjSmJqaumfPnj179uzevTsrK2vdunXPPfecd+sU3DwBAABQteTk5Pz8fGU7KipK2WjY\nsGF8fLyy3aFDB4fDMWzYsIULF4aHh3unSiEEwQ4AAKBqZrPZbDZX3cfpdAohDAZDrVR0WwQ7\nAACAGrtw4cLOnTuFEE6n89y5czNnzuzbt29wcLB3qyLYAQAA1FhGRkZGRoayHR4e3rt37/T0\ndO+WJLh5AgAAwGXBggUFBQU3b1dSUFDgdHPlypWMjIyIiIharPTWCHYAAACSINgBAABIgmAH\nAAAgCYIdAACAJAh2AAAAkiDYAQAASILn2AEAAN9gMBgc8xZ5cECNRuPB0eoCgh0AAPANXv/C\nrrqPYAcAAHyDzWZTvpLVU/R6vVYr1WVpBDsAAOAbbDbbtkyzBwfsOarCg6PVBVKlVAAAgHsZ\nwQ4AAEASBDsAAABJEOwAAAAkQbADAACQBMEOAABAEgQ7AAAASRDsAAAA/m3WrFkxMTE3b9+s\nuLj45Zdfjo2NNZlM0dHRzz333A8//FBbZd4WDygGAACoGavVGh8fX1paOn369JiYmBMnTixZ\nsqRnz567du0KDAz0YmEEOwAAgJqZP39+cXHxd999FxERIYRITk4eNGhQs2bNMjIyJk6c6MXC\nCHYAAAA14HA4li5dumDBAiXVKRo0aPDhhx96d7lOEOwAAABq5MyZM8XFxQkJCZXan3rqKa/U\n446bJwAAAGrg1KlTQoiGDRt6u5BbINgBAABUJTs7u95Pjhw50qRJEyHExYsXK3W7fPny0aNH\nvVHgzzgVCwAAUJXk5OT8/HxlOyoqSqfTBQQE5ObmxsXFuXebNGnSpUuXcnNzvVHjv7FiBwAA\nUBWz2Rz9Ez8/P51ON2XKlPnz51+6dMnV59SpUzk5Ob169fJinYJgBwAAUFNz5swJCQlp3759\nenr61q1b09PTu3btet99902bNs27hXEqFgAAoGZMJtOePXtmz569YsWKU6dONWrUqG/fvvPm\nzeNxJwAAAHXFggULFixYcPP2zYKCgpYuXVpbdVUXp2IBAAAkQbADAACQBMEOAABAElxjBwAA\nbmHs2LFVd1izZk3tVILqq1mwKyoqysvLKywsTE5ODgoKMplMOp1OpcoAAIAXnTp1qmfPnu7f\nc4+6rwbB7p133pk5c2ZpaakQ4quvvjp//vysWbOWLFkybNgw1coDAABeM2jQoFatWnm7CtRA\ndYPdp59++uKLL3bt2nX06NHK2mx8fHzDhg2ffvrp4ODg3r17q1kkAACA8Pf3Tx5p8+CAWq1s\nNxtU9+d566232rRp85e//GXw4MFKS0xMzPbt29u2bfvb3/5WtfIAAAD+TafT6T1Ko9F4+2fy\nsOqu2B08eHDGjBkGg8G90Wg0PvXUU4sWLVKhMAAAgP9QVlbmcDg8OKC/v79ki3bVDXbh4eHK\n1XWVXLhwISgoyKMlAQAA72vWrJnRaPR2Ff/Bbrebv+3hwQErfpXrwdHqguqm1McffzwjI+P6\n9evujceOHcvMzOzYsaMKhQEAAG9auXJlcXFxXl6esrLj2aUyqKS6K3YLFy6MjY2Ni4t79tln\nhRA5OTl/+9vf3n///dLS0iq+Rg0AAPioP/7xj3/4wx/8/f2bNm26bNmy7OzsTZs2vfHGG40b\nN/Z2abit6q7YNW3aNC8vr02bNm+++aYQYsGCBW+88UabNm22bdvWokULNSsEAABekJ2d/fLL\nL2/atMlmsx04cCAhISEsLOy9997zdl2oSg2eYxcTE7Np06bi4uLjx49XVFS0aNEiJCREvcoA\nAIAXVVRUdOzY0WAw9OjR48SJE/Hx8aNGjZo3b56360JVanwniNlsjo2NjY+PJ9UBACCxdu3a\nFRQUCCEeffTR77//Xgih0WjKy8u9XReqUt0VuzZt2lTx6uHDhz1RDAAAqCuee+65l19+uaSk\nRKfT7dmzZ+vWrZs3b27evLm360JVqrti1/w/NW7cuLS09P/+7//Cw8P52gkAAOQze/bsq1ev\nvvvuu+np6RqNZvny5Var9YUXXvB2XeqKjIzUaDRHjhxxb3Q4HI0aNdJoNMpK1tChQzU3mTt3\n7s2jxcTEuDoYDIaYmJi1a9eqWn91V+w2btxYqcXpdG7evHns2LFvvfWWp6sCAABe5u/v//HH\nH0dERHi7EC/IysqaPXu2a3fnzp0XL15079C5c+ff/e537i3333//LYcaOXLk5MmThRBWqzU7\nOzstLa1Ro0bJyckqVC1EjW6eqESj0TzxxBOjR4+ePXv2li1bPFgTAADwut///vcHDx48d+5c\nu3btysvLDx8+HBUVFRkZ6e26VJeYmLh+/Xr3YLdhw4bExMTc3J+fZhwWFtapU6fqjNaoUSNX\nz+7du3/99ddZWVnqBbu7/RqN5s2b79y50yOlAACAumPDhg0vvvjivn37hBC/+c1vpk+fPnLk\nyG3btnm7LtUNGDCgoKDg2LFjyq7T6fz000+HDBly9yNrNJqgoKDQ0NC7H+p27irY2Wy2Tz/9\nNDAw0FPVAACAOmLDhg3PP//8uHHjLly4sG/fvo8++mjIkCFqXyJWF4SFhfXo0SMrK0vZ/e67\n7woLC1NSUtz7OByOiv90u9FcPS0Wy0cffXTo0KFhw4apV3x1T8XefIeEw+E4duzY2bNnp06d\n6umqAACAlxUWFnbo0EGj0Xz11VePPfbY/fff371795uvuZfS0KFDly1b9uqrrwohNmzY0K9f\nv4CAAPcOmzZtMhgMrt327dvv3bv3lkMtWrRo0aJFrt1Zs2bFxsaqU7UQ1Q92ly5durkxKipq\n1KhRr732mkdL+mUajUan09Xym9ZBanwIGo1GCKHVavmE7x3qTSRm0T1F1YnkdDo9Pjh+UVhY\n2JkzZ5o3b75r167k5GSn03ngwAFVTyPWHYMGDZo4ceLx48ebN2++YcMG5Wu33HXp0mXx4sWu\nXZPJJITIzs4ePXq00rJjx45WrVoJIVJTU6dMmSKEcDqdZ8+enTFjRmFh4apVq1SqvLrBLj8/\nX6UK7oBOpwsKClJjZLsag6pGjQ9Bq9UKIQICAoxGo8cHv3cwkZSJZDab+f/xHfOtWSTUmUhK\nWOSCH/Vcu3atild/9atfffDBB9u3bz916lRycvKuXbvef//9CRMm1Fp5XhQREdG9e3dlre7c\nuXN9+vQpLCx07xAaGhofH1/pqOTkZFdeioqKUjYaNmzo6tmhQweHwzFs2LCFCxeGh4erUfmd\n3xXrRRUVFVarVY2RVUmLqikqKvL4mCaTyWQyFRcX82zxu8FECgwMNBqNFovFbve5fFJX+NYs\nEupMpJCQEIPBcOPGDf5C8Ipx48b5+fmdOHHilVdeMZvN0dHRb7/99qOPPurtumrJ0KFDV6xY\nUVZW1qdPH7PZXCnY3ZLZbDabzVX3USaz+2lcz6oq2FXzPl4hBDfGAgAgGb1eP2bMGNduZGRk\n/fr1J0yY8MEHH3ixqlozaNCgSZMmXbp06S6f13vhwgUlJjmdznPnzs2cObNv377BwcEeKrOy\nqoIdJ+MAALhn5eTkfPnll+4tTqfz+++/V26aTElJkfurpxo0aNClS5cdO3Y88cQTdzNORkZG\nRkaGsq18X1d6eronCry1qoLd119/rd4bAwCAuuy9995LSkpq1KiRq8XhcBw4cKBDhw5CCPd2\nmbjfLfrVV1+5tqOjo12XBLiehPKLCgoKPFhbddztNXZbtmxZsmTJX/7yF49UAwAA6ohHHnlk\n3LhxYWFhrha73b5///4RI0Z4sSpUrbrBzul0rl279quvviorK3Nv3Llzp3sLAACQwxtvvFGp\nRafTuT/jA3VQdYPdO++88+tf/zooKKiioqKkpKRJkyZ2u/3ChQsNGzZcunSpqiUCAIDaN3fu\n3LvsgNpX3a8UW7ly5aOPPlpYWPiPf/xDq9Xm5uaeP38+Ly/PbrcnJSWpWiIAAKh97t+XVVRU\n9M0339hsNmX3+vXr33zzjbcLxC1Ud8Xu1KlTL7zwgr+/f1RUVPv27Xfv3t24ceNOnToNHz58\n5syZf/rTn1StEgAA1DL3U7HHjh2bOHHivHnzlKdGHzlyZPLkyd4rDbdV3RW7gIAAh8OhbLdr\n12779u3KdseOHXNzc1UpDQAAADVR3WAXExPz5ZdfKvdJtG3bduPGjcpNvwUFBRaLRcUCAQAA\nUD3VPRX70ksvDR48+MEHHzx27FhSUtILL7zw3HPPtWnT5v3333/88cdVLREAAEAIYTQa7T2+\n9eCAyhdby6S6wW7gwIErV67MzMx0Op2PPvrom2++OWfOHJvN1qRJE+58BgBAbv7+/i1btvR2\nFRLmMI+rbrDTaDTjx48fP368svvKK6+88MILp0+ffuihh/z8/FQrDwAAeF90dBocv4EAACAA\nSURBVLT7V8TWr19/9OjRtV9GSUmJadd3HhywonMn5XYQaVQ3+aakpPzpT3+yWq2ulqCgoEce\neYRUBwDAvaZ+/fppaWm1/76uL/XC7VR3xW7Hjh1bt24NCgp66qmnUlNTO3fuzHIoAAASGzJk\nSNUdNmzYUDuVoPqqG+wuX7785ZdfZmZmrlu3bs2aNU2bNk1NTR01alSLFi1UrQ8AAHjF2LFj\nb2788ccfd+zYcfjwYRbP6qbqBruAgIDBgwcPHjzYarVu2bIlMzNz8eLFr7/+ekJCQmpq6oQJ\nE1StEgAA1LInnnjCtW2xWHJzc7/55pt9+/Y1adJk9OjRXbp08WJtuJ3qBjsXk8k0ZMiQIUOG\nFBUVzZgxY9WqVTt27CDYAQAgn+vXr+fm5m7btm3//v1Nmzbt1q3b5MmTmzRp4u26cFs1Dnb/\n+te/vvzyy88+++zzzz+3WCxhYWGDBw9WozIAAOBFU6dOPXjwYLNmzbp27frrX/+6UaNG3q4I\nv6y6wa6wsPDzzz//7LPPtm7dWlZWVq9evSeffHLYsGHJyckGg0HVEgEAQO07fPhwaGhoUlJS\nQkJCVFSUt8tBtVT3ztbIyMixY8d+8803w4YN27Rp0z//+c8PP/ywT58+pDoAAKS0cePGiRMn\n/uMf/5g4ceLo0aNXr179/fffS3/PRGRkpEajOXLkiHujw+Fo1KiRRqM5fPiwEGLo0KGam8yd\nO/eWAxYXF7/88suxsbEmkyk6Ovq555774Ycf1Ku/uit2w4cPHzZsWEpKitFoVK8aAABQR5jN\n5uTk5OTk5JKSkl27dn3zzTcvvvhivXr1unbt2qVLl5iYGIkffJaVlTV79mzX7s6dOy9evOje\noXPnzr/73e/cW+6///6bx7FarfHx8aWlpdOnT4+JiTlx4sSSJUt69uy5a9euwMBANSqvbrD7\n+OOP1Xh7AABQN126dMm1HRMTExMTk5aWtmfPnm+++WbdunXh4eFZWVleLE89iYmJ69evdw92\nGzZsSExMzM3NdbWEhYV16tTpF4eaP39+cXHxd999FxERIYRITk4eNGhQs2bNMjIyJk6cqEbx\nNb55AgAA3AueeeaZKl69evVqrVVSywYMGPDqq68eO3bsoYceEkI4nc5PP/30xRdfdA921eFw\nOJYuXbpgwQIl1SkaNGjw4YcfqrRcJwh2AADgltasWePtErwjLCysR48eWVlZr776qhDiu+++\nKywsTElJce/jcDgqKircW/T6ypnqzJkzxcXFCQkJldqfeuopFar+qQz1hgYAAL7rgQce8HYJ\nXjN06NBly5YpwW7Dhg39+vULCAhw77Bp0yb3+0fbt2+/d+/eSoOcOnVKCNGwYUP16/2ZtJc9\nAgAA3JlBgwYdPnz4+PHjTqdzw4YNQ4cOrdShS5cue9z88Y9/FEJkZ2fX+8mRI0eUJzlXuutC\nCHH58uWjR4+qVDkrdgAAAP8hIiKie/fuylrduXPn+vTpU1hY6N4hNDQ0Pj6+0lHJycn5+fnK\ndlRUlE6nCwgIyM3NjYuLc+82adKkS5cu1fSKvWpixQ4AAKCyoUOHrl+/fsOGDX369DGbzdU5\nxGw2R//Ez89Pp9NNmTJl/vz57vcXnzp1Kicnp1evXiqVTbADAACobNCgQfn5+StWrLj5PGz1\nzZkzJyQkpH379unp6Vu3bk1PT+/atet99903bdo0D5bqjmAHAABQWYMGDbp06XLlypUnnnji\njgcxmUx79uwZNmzYihUrBg4cuGzZsr59++bl5fG4EwAAANW5nzb96quvXNvR0dGur1Or0ZOZ\ng4KCli5d6qnyfhErdgAAAJIg2AEAAEiCYAcAACAJgh0AAIAkCHYAAACSINgBAABIgmAHAAAg\nCZ5jBwAAfIPJZHJ0TfDggBqNxoOj1QWs2AEAAN/gekQwbocVOwAA4BtKSkpe+LPJgwOuTLPr\ndDoPDuh1rNgBAABIgmAHAAAgCYIdAACAJAh2AAAAkiDYAQAASIJgBwAAIAmCHQAAgCQIdgAA\nAP82dOhQjZsmTZpMmTLlxx9/dHUoKyt7/fXXExMTg4ODH3rooeHDhx8+fLj6h7uLiYlx9TQY\nDDExMWvXrr3L+nlAMQAAwM8SEhIWL14shLDZbPv27fuf//mfkpKSVatWCSGuXLmSnJx84cKF\nadOmvfbaaxcvXszMzGzfvv3KlStTU1N/8fCbjRw5cvLkyUIIq9WanZ2dlpbWqFGj5OTkOy6e\nYAcAAPCz0NDQTp06KdudO3e+cuXKsmXLVq5cqdFoZs6cef369QMHDkRFRSkd0tLSZs2aNXny\n5OTkZKWxisNvfq9GjRq5Onfv3v3rr7/Oysq6m2DHqVgAAIDbqlevntVqtdvtP/zww5o1a377\n29+6Up0QQqPRzJ07NyQk5O2336768F98I41GExQUFBoaejfVsmIHAADwM4fDUVFRIYSoqKjY\nv3//O++8k5SUpNfrlWvpUlJSKvUPCAjo0qXLwYMHqz686vdSTsUeOnTo3XffvZviCXYAAAA/\n27Jli8FgcO22bdt29erVQojTp08HBARERETcfEjTpk337dtX9eG3tGjRokWLFrl2Z82aFRsb\nezfFE+wAAAB+lpiYmJ6ermyHhYVFR0drtVohRHR0dElJydWrV8PDwysdcu7cuRYtWlR9eHZ2\n9ujRo5X2HTt2tGrVSgiRmpo6ZcoUIYTT6Tx79uyMGTMKCwtvd6dFdRDsAAAAflavXr34+Pib\n29u0aSOEyMnJGTFihHt7aWnptm3bxowZU/XhycnJ+fn5yrbrKr2GDRu6Onfo0MHhcAwbNmzh\nwoU3Z8dqItgBAAD8sqioqLFjx77yyivdu3dv2LCh0uh0OufMmXPx4sVnn3226sPNZrPZbK66\nj9PpFEK4n8mtKYIdAABAtSxcuLBHjx5t27adPn16bGzspUuX1q1bt2XLlg8++KBly5Z3MOCF\nCxd27twphHA6nefOnZs5c2bfvn2Dg4PvuEKCHQAAQLVERETs3r174cKF2dnZb7zxRlhYWPv2\n7fft2xcXF3dnA2ZkZGRkZCjb4eHhvXv3dl2fd2cIdgAAAP+WlZVVdQd/f//Zs2fPnj37zg53\nV1BQUIPKqocHFAMAAEiCYAcAACAJgh0AAIAkCHYAAACSINgBAABIgmAHAAAgCR53AgAAfING\nozH7e7uIuo1gBwAAfENAQMC7Iz07pM6zw3kdwQ4AAPgGu93u2QG1Wq1Go/HsmN5FsAMAAL6h\nrKxMvHjNgwP6v99Qp5Nq0Y6bJwAAACRBsAMAAJAEwQ4AAEASBDsAAABJ1EawczgcH3/88bhx\n49LS0latWlXFLS12u/3ZZ5+9fv16LVQFAAAgmdq4KzYzM/OLL7544YUX9Hr98uXLtVrt2LFj\nb+5WXl7+ySef/Pjjj7VQEgAAgHxUD3YVFRWbN28eNWpUQkKCEKKsrGz58uUjRowwGo3u3TZt\n2vThhx9WVFSoXQ8AAICsVD8Ve/78+aKiori4OGU3Li7OarWeOHGiUrcuXbosXbp0xowZatcD\nAABwO0OHDtW4adKkyZQpU9xPJ5aVlb3++uuJiYnBwcEPPfTQ8OHDDx8+XP3DKykuLn755Zdj\nY2NNJlN0dPRzzz33ww8/3E39qq/YXbt2TQgRHh6u7JrNZqPRWFRUVKlbSEhISEiIzWa75SAW\ni2Xy5Mmu3UGDBg0cOFCNaj38QGuV1atXz+NjarVaIYTZbDaZTB4f/N7BRFImUnBwsNPp9Pjg\n9wjfmkVCnYmkPDk2JCTE4yNDcePGDW+XUBclJCQsXrxYCGGz2fbt2/c///M/JSUlq1atEkJc\nuXIlOTn5woUL06ZNe+211y5evJiZmdm+ffuVK1empqb+4uGVWK3W+Pj40tLS6dOnx8TEnDhx\nYsmSJT179ty1a1dgYOCdFa96sLNYLAaDQa//+Y1MJlNNZ5Ldbj969Khrt3Pnzu4DepBv/SZV\n6UMQP/0yxR1jIimYSHfDt2aRUHMiqTcycEuhoaGdOnVStjt37nzlypVly5atXLlSo9HMnDnz\n+vXrBw4ciIqKUjqkpaXNmjVr8uTJycnJSmMVh1d6o/nz5xcXF3/33XcRERFCiOTk5EGDBjVr\n1iwjI2PixIl3Vrzq/1oCAwNtNpvdbnf9irdarTXNoaGhoXv37nXtWq3WK1eueLLKnwSpMahq\n1PgQTCaTyWSyWCzl5eUeH/zewUQKDAw0Go3Xr1/3+Bc73jt8axYJdSZSSEiIwWC4evUqS7/w\nonr16lmtVrvdfvny5TVr1mRkZLhSnRBCo9HMnTv3448/fvvttxcuXFjF4ZX+RHE4HEuXLl2w\nYIGS6hQNGjT48MMP73i5TtRCsAsNDRVCXLt2rX79+kKIkpKS0tJSpREAAKCucTgcyt2cFRUV\n+/fvf+edd5KSkvR6vXItXUpKSqX+AQEBXbp0OXjwYNWHVzrqzJkzxcXFyq2l7p566qm7KV71\nmyeio6NDQkLy8/OV3fz8/ICAgBYtWqj9vgAAAHdgy5YtBoPBYDAEBAQkJCSEhYWtXr1aCHH6\n9OmAgAD3BTaXpk2bnjx5surDKzl16pQQomHDhp4tXvUVO51O17dv34yMjMjISK1Wu3r16l69\neinPOsnJySkrKxswYIDaNQAAAFRTYmJienq6sh0WFhYdHa3cEBYdHV1SUnL16lXXLaEu586d\ncy1a3e7w7Ozs0aNHK+07duxo0qSJEOLixYuVst3ly5evXr368MMP31nxtXFF6vDhw202W3p6\nusPhSEpKSktLU9rz8vIsFgvBDgAA1B316tWLj4+/ub1NmzZCiJycnBEjRri3l5aWbtu2bcyY\nMVUfnpyc7DqBGRUVpdPpAgICcnNzXY+EU0yaNOnSpUu5ubl3VnxtBDuNRpOamuq6Ddhl7ty5\nlVqaN2/+v//7v7VQEgAAQI1ERUWNHTv2lVde6d69u2uZzel0zpkz5+LFi88++2zVh5vNZrPZ\n7N4yZcqU+fPnDxs2LDIyUmk5depUTk7O3TzWl3vIAQAAqmXhwoU9evRo27bt9OnTY2NjL126\ntG7dui1btnzwwQctW7as6Whz5szJzs5u3779jBkzWrVqdeTIkSVLltx3333Tpk274woJdgAA\nANUSERGxe/fuhQsXZmdnv/HGG2FhYe3bt9+3b1+l06nVZDKZ9uzZM3v27BUrVpw6dapRo0Z9\n+/adN29enX7cCQAAgK/IysqquoO/v//s2bNnz559Z4dXEhQUtHTp0hodUjXVH3cCAACA2kGw\nAwAAkATBDgAAQBIEOwAAAEkQ7AAAACRBsAMAAJAEjzsBAAC+QafT2d8O83YVdRrBDgAA+AZ/\nf39vl1DXcSoWAABAEgQ7AAAASRDsAAAAJEGwAwAAkATBDgAAQBIEOwAAAEkQ7AAAACRBsAMA\nAJAEwQ4AAEASBDsAAABJEOwAAAAkQbADAACQhN7bBeAe8nJ2hBrD/m7gFTWGBQDA57BiBwAA\nIAmCHQAAgCQ4FQufZ3qlVI1hrfONagwLAIB6WLEDAACQBMEOAABAEgQ7AAAASRDsAAAAJEGw\nAwAAkATBDgAAQBIEOwAAAEnwHDsAwD1HpW84FHzJIbyNFTsAAABJsGIHwMeotNbCQgs8gu/C\ngXcR7ABACP5/DEAKnIoFAACQBMEOAABAEgQ7AAAASRDsAAAAJEGwAwAAkATBDgAAQBIEOwAA\nAEkQ7AAAACRBsAMAAJAEwQ4AAEASBDsAAABJEOwAAAAkQbADAACQBMEOAABAEgQ7AAAASRDs\nAAAAJEGwAwAAkATBDgAAQBIEOwAAAEkQ7AAAACRBsAMAAJAEwQ4AAEASBDsAAABJEOwAAAAk\nQbADAACQBMEOAABAEgQ7AAAASRDsAAAAJEGwAwAAkATBDgAAQBIEOwAAAEkQ7AAAACRBsAMA\nAJAEwQ4AAEASBDsAAABJEOwAAAAkofd2AXdCp9MFBgZ6uwrvU+ND0Ov1Qgij0ejn5+fxwX3L\nvTPH1PhJDQaDEMJkMjmdTo8P7luYSHdDp9MJIcxms8dH9jkqTaQbN26oMSy8yCeDncPhKC8v\nV2NkoxqDqqasrEyNYfV6vc1mq6ioUGFsX/qA7+bj9aWfU52JpNVqdTqdzWaz2+0eH9y3PuA7\n/nh96YcUQqgzkfR6vVarLS8vV+EvBB/7gFX6hQ/5+GSwczqdNptNjZF96x+6Gh+CstBit9tV\n+oR9yN18Akwkf39/ZWR1gp0vueOP17dmkVBnIil5zmazsfTL72RUE9fYAQAASIJgBwAAIAmC\nHQAAgCQIdgAAAJIg2AEAAEiCYAcAACAJgh0AAIAkCHYAAACSINgBAABIgmAHAAAgCYIdAACA\nJAh2AAAAkiDYAQAASIJgBwAAIAmCHQAAgCQIdgAAAJIg2AEAAEiCYAcAACAJgh0AAIAkCHYA\nAACSINgBAABIgmAHAAAgCYIdAACAJAh2AAAAkiDYAQAASIJgBwAAIAmCHQAAgCQIdgAAAJIg\n2AEAAEiCYAcAACAJgh0AAIAkCHYAAACSINgBAABIgmAHAAAgCYIdAACAJAh2AAAAkiDYAQAA\nSIJgBwAAIAmCHQAAgCQIdgAAAJIg2AEAAEiCYAcAACAJgh0AAIAkCHYAAACSINgBAABIgmAH\nAAAgCYIdAACAJAh2AAAAkiDYAQAASIJgBwAAIAmCHQAAgCQIdgAAAJIg2AEAAEiCYAcAACAJ\ngh0AAIAkCHYAAACSINgBAABIgmAHAAAgCYIdAACAJAh2AAAAkiDYAQAASIJgBwAAIAmCHQAA\ngCQIdgAAAJIg2AEAAEiCYAcAACAJgh0AAIAkCHYAAACSINgBAABIgmAHAAAgCYIdAACAJAh2\nAAAAkiDYAQAASIJgBwAAIAl9LbyHw+H485///Pe//91utyclJY0ZM0an091BHwAAAFShNoJd\nZmbmF1988cILL+j1+uXLl2u12rFjx95BHwAAAFRB9WBXUVGxefPmUaNGJSQkCCHKysqWL18+\nYsQIo9FYoz4AAAComurX2J0/f76oqCguLk7ZjYuLs1qtJ06cqGkfAAAAVE31Fbtr164JIcLD\nw5Vds9lsNBqLiopq1Ke8vHzbtm2u3fvvv79Ro0ZqV173+fv7e3xM5dJGg8Gg0Wg8PrhvUePj\nrZvUm0h+fn4Oh8Pjg/sWJtLd0Gq1yshOp9Pjg/uWe2ci4S6pHuwsFovBYNDrf34jk8l048aN\nGvUpLi6eNWuWa/f5559//vnn1ai2TI1BVRO8s5cq4xreUmVYIcaIRJVGVkNQUNAdH8tEEoKJ\n9G93PJF8axYJX5tIvjWLxN39RqrC5cuX1RgWXqR6sAsMDLTZbHa73XWXq9VqDQwMrFGfgICA\nKVOmuHZbt25dXFysSrmz56sxamehyrD/UmFMg8Hg5+dXWlpqt9s9P3oHVf6rOYeHqjHsXc2x\ne34i+fv76/X6kpISVVbs7pGJpM4sEj41kYxGo06nU+UXvjqzSNTBiYR7jOrBLjQ0VAhx7dq1\n+vXrCyFKSkpKS0uVxur3MRqNo0ePdu1arVar1ap25fcmjUbj5+dXXl5eXl7u7Vrgw3Q6nV6v\nV+svBNwz/Pz8dDpdaWkpp2KBalL95ono6OiQkJD8/HxlNz8/PyAgoEWLFjXtAwAAgKqpvmKn\n0+n69u2bkZERGRmp1WpXr17dq1cv5TkmOTk5ZWVlAwYMqKIPAAAAqqk2HlA8fPhwm82Wnp7u\ncDiSkpLS0tKU9ry8PIvFMmDAgCr6AAAAoJo0vnjhAtfYqcdkMplMJovFwjV2uBuBgYFGo/H6\n9etcY4e7ERISYjAYrl696ov/q/IJN27cUOkfacuWLdUYFr9I9WvsAAAAUDsIdgAAAJIg2AEA\nAEiCYAcAACAJgh0AAIAkCHYAAACSINgBAABIgmAHAAAgCYIdAACAJAh2AAAAkiDYAQAASIJg\nBwAAIAmCHQAAgCQIdgAAAJIg2AEAAEiCYAcAACAJgh0AAIAk9N4uAHWLxWK5cuWK0WjUagn9\nuHPXr1+32WxMJNylwsLCiooKk8nk7UIAn6FxOp3ergF1yOrVq3//+98vXbq0c+fO3q4FPuz1\n11/Pzs7OzMx88MEHvV0LfNikSZN27969ffv2gIAAb9cC+Ab+mAYAAJAEwQ4AAEASBDsAAABJ\ncPME/kNSUlJoaGiLFi28XQh82xNPPNG6deuIiAhvFwLfNmLEiOTkZD8/P28XAvgMbp4AAACQ\nBKdiAQAAJEGwAwAAkATX2OHfFixYsGPHjkqNzzzzzDPPPOOVeuBzUlNTi4qKli9f3rhxY1ej\n0+kcM2bMtWvX3n333aZNm3qxPNRlY8aM0ev1y5Yt8/f3dzW++OKLCQkJTz/9tKulsLBw3Lhx\n/v7+H330kdFo9EalQF1HsMPPWrdunZaW5t4SHh7upVrgq7799tvhw4e7do8dO3b9+nUv1gNf\n8c9//nPdunWpqalV9Nm+fbufn19ZWdmePXt4iDpwS5yKxc8CAwMf+k/c1Ygaefjhh7/99lv3\nlh07djz88MPeqgc+5NFHH/3ss8/Onj1bRZ9t27Z16tSpefPmubm5tVYY4FsIdgA85rHHHjt/\n/vyFCxeUXafTuWPHjoSEBO9WBZ+QmJjYtm3b5cuX3+5ZDefPnz958mRiYmLHjh337t1rtVpr\nuULAJxDs8DOn02n/T96uCD4mMDCwbdu2rkW7EydOWCyWdu3aebcq+IqJEyeeOHHiL3/5yy1f\n3b59u7+/f1xcXMeOHW02265du2q5PMAnEOzws927dw92M2PGDG9XBN+TkJDgCnY7duzo0KGD\n++XwQBUiIyOffvrpP/zhDzdu3Kj0ktPp3LZtW3x8vL+//wMPPBAREbF9+3avFAnUcdw8gZ+1\nbt163Lhxrl3+f4w70KlTp/fee++HH35o2LBhXl7eyJEjvV0RfMngwYO//vrrNWvWTJ061b39\n5MmTFy5cGDFihBBCo9F07NgxJyfnX//6V2BgoJcqBeoogh1+FhgY2Lx5c29XAd8WHBz8yCOP\nKGt1hYWF7du3t1gs3i4KPkOv10+aNOmVV15JTk52b1fW5956663FixcLIZxOp9Pp3LlzZ6Vu\nAAh2ADwsMTExJyfHZrO1b9/eaDQS7FAjrVu3Tk5Ofu+991wtynnYdu3ajR071tU4e/bsbdu2\nEeyASrjGDoCHderU6eTJkzk5OYmJid6uBT4pLS3NYrG4bq8+evTolStX+vXr19RNjx49Dh48\nePPVeMA9jmAHwMPq1avXpk0bi8USHx/v7Vrgk4KDg90X57Zv3x4eHh4XF+fep2fPng6H4+bv\nywHucZrbPTEIAAAAvoUVOwAAAEkQ7AAAACRBsAMAAJAEwQ4AAEASBDsAAABJEOwAAAAkQbAD\nAACQBMEOAABAEgQ7AAAASRDsANS2t956S6PRXLlyxduFAIBsCHYAAACSINgBAABIgmAHAAAg\nCYIdgH/Lycnp1q1bSEjI448/npmZ+dZbbxmNRterZ8+eHTFiRHR0dFBQUFJS0meffeZ6qXfv\n3oMGDSooKOjZs6fZbI6MjBw/fvyNGzdcHT755JOEhITg4OD27dsvW7as0vtWMXJycvLQoUOP\nHz/eu3fv6OhotX5yAJCF3tsFAKgTMjMzn3nmmdatW0+dOvWf//xnWlrao48+6nq1oKAgISHB\nZDKNHDnSaDRmZWU9+eST77zzzpQpU5QOFy9e7N69++DBg4cNG/bXv/519erVTqdz9erVQogl\nS5ZMnz69RYsWkydPLioqmjlz5n333Vf9kYuKivr376/X63v37l2LnwcA+CYngHteaWlp48aN\n27VrV1xcrLRs2bJFCOHv76/s9u/fv2nTpteuXVN2y8vLu3btGhAQcP36dafTmZKSIoR4++23\nlVcdDkfHjh0bN27sdDoLCwuDgoLatm1rsViUV3fu3KnRaIQQhYWFvzhyjx49hBATJkxwOBy1\n81EAgE/jVCwAsXPnznPnzk2bNs1kMiktKSkpbdq0UbaLi4s3bdo0cuRIjUZTVFRUVFRUXFw8\nevTokpKSvLw8pY+fn9+ECROUbY1G07ZtW6vVKoT4+uuvf/zxx//3//5fUFCQ8upjjz3Wp0+f\n6o+s0WgWL16sZEEAQNUIdgDE8ePHhRCuJCeE0Gg0rVq1cn/1zTffDHUzduxYIURhYaHSp0mT\nJv7+/q7DtVqt+7Ht2rVzfzvXSd7qjBwdHW02mz39EwOAnLjGDoCw2Ww3N7rCWUVFhRBi1qxZ\nyilXdy1btlQ23FOdO73+Fr9kajRyWFhYNX4CAIAQBDsA4qcUdeTIkdjYWFfj0aNHlY0WLVoI\nIfR6fbdu3VyvFhQU7N27t3379lWP3KxZMyHE/v37lUEUBw8evPuRAQA341QsANGhQ4eIiIjF\nixeXlJQoLVu3bj1w4ICyHRIS0rVr1/fff//YsWNKS1lZWWpq6syZM13X5N2O8vyUN99802Kx\nKC27d+/+4osv7n5kAMDNWLEDIIKDgxcsWDB+/PhOnTo9+eSTly9fXr9+fXx8fEFBgdJhyZIl\nnTt3TkxMfPrppxs1arR+/fr8/PxPPvlEp9NVPXJYWNhvfvObX//61/Hx8UOGDLFYLGvXru3c\nufO2bdvucmQAwM1YsQMghBDjxo3buHFjQEDAkiVLjhw5snHjxhYtWrgeOBcXF7d///7OnTtn\nZ2cvXLgwMDBw8+bNTz/9dHVGfvHFF9etW1e/fv3ly5fn5eX99re/feWVV1yv3s3IAIBKNE6n\n09s1APCysrKyQ4cONW7cuEGDBq7G7t27l5aWuh47AgCo+1ixAyA0Gk3Xrl1nzpzpajl79mxu\nbu6vfvUrL1YFAKgprrEDIPz8/KZMmbJw4UK9Xt+jR4/z58+//fbbZrPZ9b1eAACfwKlYAEII\nUVFR8e67765aterUqVNRUVGPPfbYggULGjdu7O26AAA1QLADAACQBNfYgHDrYAAAAD1JREFU\nAQAASIJgBwAAIAmCHQAAgCQIdgAAAJIg2AEAAEiCYAcAACAJgh0AAIAkCHYAAACSINgBAABI\n4v8DOLMw24UUPvIAAAAASUVORK5CYII=",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "ggplot(combined.tall %>% filter((DataSet == \"LastFM\" | DataSet == \"MovieLens-1M\"))) +\n",
    "    aes(x=gender, y=value, fill=Algorithm) +\n",
    "    geom_bar(stat=\"identity\", position=\"dodge\") +\n",
    "    facet_grid(DataSet ~., scales=\"free\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analyze by Age"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using DataSet, Algorithm, age as id variables\n"
     ]
    }
   ],
   "source": [
    "age.results = user.results %>%\n",
    "    group_by(DataSet, Algorithm, age) %>%\n",
    "   summarize(MAP=mean(AvgPrec), MRR=mean(RecipRank),\n",
    "             \n",
    "              nDCG=mean(TopN.nDCG))\n",
    "age.tall = melt(age.results, variable.name = \"Metric\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using DataSet, Algorithm, age as id variables\n"
     ]
    }
   ],
   "source": [
    "overall.results = user.results %>%\n",
    "    group_by(DataSet, Algorithm) %>%\n",
    "    summarize(MAP=mean(AvgPrec), MRR=mean(RecipRank),\n",
    "              \n",
    "              nDCG=mean(TopN.nDCG))\n",
    "combined.results = rbind(age.results, mutate(overall.results))\n",
    "combined.tall = melt(combined.results, variable.name='Metric')\n",
    "combined.tall$Algorithm[combined.tall$Algorithm == \"Pop\"] <- \"POP-B\"\n",
    "combined.tall$Algorithm[combined.tall$Algorithm == \"PopCt\"] <- \"POP-C\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Warning message:\n",
      "“Removed 24 rows containing missing values (geom_bar).”"
     ]
    },
    {
     "data": {},
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0gAAAJYCAIAAADXJFGjAAAACXBIWXMAABJ0AAASdAHeZh94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\nrkJggg==",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ggplot(combined.tall %>% filter( DataSet==c(\"LastFM\",\"Lenskit1M\"),\n",
    "                                grepl(\"(-B)|(-C)\", Algorithm)  )) +\n",
    "    aes(x=age, y=value, fill=Algorithm) +\n",
    "    geom_bar(stat=\"identity\", position=\"dodge\") +\n",
    "    facet_grid(Metric ~ DataSet, scales=\"free\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "R [conda env:my-r-env]",
   "language": "R",
   "name": "conda-env-my-r-env-r"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "3.3.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}