BICA_BIMAN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## This Script has code for Running the BICA and the final BIMAN model\n",
    "\n",
    "Following files are assumed present:\n",
    "1. Bot data content\n",
    "2. Non-Bot data content\n",
    "3. Result of BIN \n",
    "4. Result of BIM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Loading required package: Formula\n",
      "Loading required package: plotrix\n",
      "Loading required package: TeachingDemos\n"
     ]
    }
   ],
   "source": [
    "library(data.table)\n",
    "library(randomForest)\n",
    "library(AUC)\n",
    "library(plotmo)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Bot Data Content\n",
    "botdata = fread('/da4_data/play/botDetection/bot_Info.csv.gz', header = T)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classes 'data.table' and 'data.frame':\t13150 obs. of  19 variables:\n",
      " $ Author                : chr  \"010011110110000101101011 <bot@oakmade.com>\" \"0asa <vince.bot@gmail.com>\" \"0q0.eu Team <bot@0q0.eu>\" \"0x48piraj [BOT] <31832371+5p1d3r@users.noreply.github.com>\" ...\n",
      " $ No.Commit             : int  1 91 5 20 3 1 337 2 34 4 ...\n",
      " $ Days.Active           : int  1 381 1 1 1 1 159 3 790 1 ...\n",
      " $ Avg.Commit.pYear      : num  1 45.5 5 20 3 ...\n",
      " $ Median.Commit.pYear   : num  1 45.5 5 20 3 ...\n",
      " $ Activity.Hours        : int  1 15 4 2 2 1 22 2 7 2 ...\n",
      " $ Spike.Hours           : int  1 2 4 2 2 1 8 2 1 1 ...\n",
      " $ Circ.Variance.Hour    : num  0 3.1213 0.1892 0.0201 0.2612 ...\n",
      " $ Tot.uniq.FilesChanged : int  1 108 91 3 83 1 7151 1 5 2 ...\n",
      " $ Tot.FilesChanged      : int  1 210 97 20 83 1 9140 2 34 4 ...\n",
      " $ Uniq.File.Exten       : int  1 29 3 3 31 1 240 1 2 2 ...\n",
      " $ Avg.File.pCommit      : num  1 2.31 19.4 1 27.67 ...\n",
      " $ Std.File.pCommit      : num  0 2.04 38.91 0 46.19 ...\n",
      " $ No.Timezones          : int  1 3 1 2 1 1 2 1 2 1 ...\n",
      " $ Std.Timezones         : num  0 1.25 0 4.44 0 ...\n",
      " $ Tot.Projects          : int  1 5614 5 20 3 74 10090 4 400 17 ...\n",
      " $ Tot.uniq.Projects     : int  1 2746 1 1 1 74 65 2 371 15 ...\n",
      " $ Median.Project.pCommit: num  1 2 1 1 1 74 25 2 1 1 ...\n",
      " $ Std.Project.pCommit   : num  0 350 0 0 0 ...\n",
      " - attr(*, \".internal.selfref\")=<externalptr> \n"
     ]
    }
   ],
   "source": [
    "str(botdata)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classes 'data.table' and 'data.frame':\t13150 obs. of  19 variables:\n",
      " $ Author                : chr  \"adiel <adam@adamdiel.com>\" \"YannDoy <doy.yann@hotmail.com>\" \"irvincito <isc.irving.rosales@gmail.com>\" \"Jesper Corba <jesper@discworld.abf.nl>\" ...\n",
      " $ No.Commit             : int  6 43 1 2 87 1 1 1 96 1 ...\n",
      " $ Days.Active           : int  5 1002 1 1 1425 1 1 1 1732 1 ...\n",
      " $ Avg.Commit.pYear      : num  6 14.3 1 2 29 ...\n",
      " $ Median.Commit.pYear   : num  6 2 1 2 7 1 1 1 11 1 ...\n",
      " $ Activity.Hours        : int  2 6 1 1 17 1 1 1 18 1 ...\n",
      " $ Spike.Hours           : int  2 2 1 1 4 1 1 1 4 1 ...\n",
      " $ Circ.Variance.Hour    : num  1.231 0.561 0 0 3.808 ...\n",
      " $ Tot.uniq.FilesChanged : int  15 19 18 2 50 1 2 31 304 1 ...\n",
      " $ Tot.FilesChanged      : int  17 69 18 2 179 1 2 31 395 1 ...\n",
      " $ Uniq.File.Exten       : int  10 7 12 2 11 1 2 11 59 1 ...\n",
      " $ Avg.File.pCommit      : num  2.83 1.6 0 1 2.06 ...\n",
      " $ Std.File.pCommit      : num  2.71 1.26 0 0 1.79 ...\n",
      " $ No.Timezones          : int  1 2 1 1 2 1 1 1 3 1 ...\n",
      " $ Std.Timezones         : num  0 0.355 0 0 0.693 ...\n",
      " $ Tot.Projects          : int  6 43 1 2 156 1 1 1 122 1 ...\n",
      " $ Tot.uniq.Projects     : int  1 3 1 1 5 1 1 1 9 1 ...\n",
      " $ Median.Project.pCommit: num  1 1 0 1 2 0 0 0 1 0 ...\n",
      " $ Std.Project.pCommit   : num  0 0 0 0 0.407 ...\n",
      " - attr(*, \".internal.selfref\")=<externalptr> \n"
     ]
    }
   ],
   "source": [
    "# Bot Data Content\n",
    "nonbotdata = fread('/da4_data/play/botDetection/non_bot_Info.csv.gz', header = T)\n",
    "str(nonbotdata)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "botdata$isBot = 1\n",
    "nonbotdata$isBot = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "   No.Commit          Days.Active      Avg.Commit.pYear   Median.Commit.pYear\n",
       " Min.   :      1.0   Min.   :    1.0   Min.   :     1.0   Min.   :     1.0   \n",
       " 1st Qu.:      1.0   1st Qu.:    1.0   1st Qu.:     1.0   1st Qu.:     1.0   \n",
       " Median :      3.0   Median :    2.0   Median :     3.0   Median :     3.0   \n",
       " Mean   :    528.2   Mean   :  185.1   Mean   :   159.0   Mean   :   152.4   \n",
       " 3rd Qu.:     17.0   3rd Qu.:  143.0   3rd Qu.:    11.5   3rd Qu.:    11.0   \n",
       " Max.   :2463758.0   Max.   :24856.0   Max.   :615939.5   Max.   :701464.0   \n",
       " Activity.Hours    Spike.Hours     Circ.Variance.Hour Tot.uniq.FilesChanged\n",
       " Min.   : 1.000   Min.   : 0.000   Min.   : 0.0000    Min.   :      0      \n",
       " 1st Qu.: 1.000   1st Qu.: 1.000   1st Qu.: 0.0000    1st Qu.:      1      \n",
       " Median : 2.000   Median : 1.000   Median : 0.0000    Median :     10      \n",
       " Mean   : 5.017   Mean   : 2.408   Mean   : 0.7988    Mean   :   1376      \n",
       " 3rd Qu.: 7.000   3rd Qu.: 3.000   3rd Qu.: 1.2313    3rd Qu.:     88      \n",
       " Max.   :24.000   Max.   :24.000   Max.   :13.3228    Max.   :5118357      \n",
       " Tot.FilesChanged   Uniq.File.Exten    Avg.File.pCommit    Std.File.pCommit   \n",
       " Min.   :       0   Min.   :    0.00   Min.   :     0.00   Min.   :     0.00  \n",
       " 1st Qu.:       3   1st Qu.:    1.00   1st Qu.:     1.00   1st Qu.:     0.00  \n",
       " Median :      20   Median :    2.00   Median :     2.00   Median :     0.00  \n",
       " Mean   :    6656   Mean   :   38.15   Mean   :    62.54   Mean   :    75.02  \n",
       " 3rd Qu.:     142   3rd Qu.:    9.00   3rd Qu.:    12.20   3rd Qu.:     5.58  \n",
       " Max.   :27335682   Max.   :50122.00   Max.   :101565.33   Max.   :175868.69  \n",
       "  No.Timezones    Std.Timezones      Tot.Projects      Tot.uniq.Projects \n",
       " Min.   : 1.000   Min.   : 0.0000   Min.   :       0   Min.   :    0.00  \n",
       " 1st Qu.: 1.000   1st Qu.: 0.0000   1st Qu.:       2   1st Qu.:    1.00  \n",
       " Median : 1.000   Median : 0.0000   Median :       5   Median :    2.00  \n",
       " Mean   : 1.309   Mean   : 0.3054   Mean   :    5299   Mean   :   75.89  \n",
       " 3rd Qu.: 1.000   3rd Qu.: 0.0000   3rd Qu.:      24   3rd Qu.:    4.00  \n",
       " Max.   :30.000   Max.   :10.1559   Max.   :46112666   Max.   :81022.00  \n",
       " Median.Project.pCommit Std.Project.pCommit isBot    \n",
       " Min.   :   0.000       Min.   :    0.000   0:13150  \n",
       " 1st Qu.:   1.000       1st Qu.:    0.000   1:13150  \n",
       " Median :   1.000       Median :    0.000            \n",
       " Mean   :   4.045       Mean   :    9.298            \n",
       " 3rd Qu.:   1.000       3rd Qu.:    0.000            \n",
       " Max.   :4667.000       Max.   :18093.728            "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Combine Data\n",
    "alldata = rbind(botdata, nonbotdata)\n",
    "all.author = alldata$Author\n",
    "alldata$Author = NULL\n",
    "alldata$isBot = factor(alldata$isBot)\n",
    "summary(alldata)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating data for Random Forest Model: 70-30 ration for Train and Test set\n",
    "ind = 1:nrow(alldata)\n",
    "indTRAIN = sample(ind,round(0.7*length(ind)))\n",
    "indTEST <- ind[-indTRAIN]\n",
    "isBot = alldata$isBot\n",
    "alldata$isBot = NULL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Subset the data based on previous iterations to get optimal result\n",
    "alldata.sub = alldata[,c(10,9,12,16,11,17)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "0.895703802962006"
      ],
      "text/latex": [
       "0.895703802962006"
      ],
      "text/markdown": [
       "0.895703802962006"
      ],
      "text/plain": [
       "[1] 0.8957038"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0gAAANICAMAAADKOT/pAAADAFBMVEUAAAABAQECAgIDAwME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",
      "text/plain": [
       "Plot with title \"Random Forest Model: Bot Detection\""
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "set.seed(10)\n",
    "BICA <- randomForest(x=alldata.sub[indTRAIN,],\n",
    "                   y=isBot[indTRAIN], ntree=100, importance=T)\n",
    "pred <- predict(BICA, alldata.sub[indTEST,], type='prob')[,2]\n",
    "auc(roc(pred,isBot[indTEST]))\n",
    "# auc(accuracy(pred,isBot[indTEST]))\n",
    "varImpPlot(BICA, type=1, main= 'Random Forest Model: Bot Detection')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Saving the Random Forest Model\n",
    "save(BICA, file='BICA_model.Rdata')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculating apartdep for Uniq.File.Exten \n",
      "calculating apartdep for Tot.FilesChanged \n",
      "calculating apartdep for Std.File.pCommit \n",
      "calculating apartdep for Tot.uniq.Projects \n",
      "calculating apartdep for Avg.File.pCommit \n",
      "calculating apartdep for Median.Project.pCommit \n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0gAAANICAMAAADKOT/pAAADAFBMVEUAAAABAQECAgIDAwME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",
      "text/plain": [
       "Plot with title \"6  Median.Project.pCmmt\""
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plotmo(BICA, type='prob', pmethod='apartdep', degree2 = F, persp.ticktype=\"detailed\", persp.nticks=3, \n",
    "       caption = 'Partial Dependence plot for all predictors in BICA model', cex = 1.3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## This concludes the BICA Training part\n",
    "## Example of using BICA to predict bots is shown below"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "ftestdata = fread('/da4_data/play/botDetection/test_Info_paper.csv.gz', header = T)\n",
    "auth = ftestdata$Author\n",
    "ftestdata$Author = NULL\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>auth</th><th scope=col>pcl</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td><span style=white-space:pre-wrap>dependabot[bot] &lt;support@dependabot.com&gt;          </span></td><td>1                                                                                               </td></tr>\n",
       "\t<tr><td><span style=white-space:pre-wrap>greenkeeper &lt;felixhennigfh@gmail.com&gt;             </span></td><td>0                                                                                               </td></tr>\n",
       "\t<tr><td><span style=white-space:pre-wrap>Greenkeeper &lt;support@greenkeeper.io&gt;              </span></td><td>1                                                                                               </td></tr>\n",
       "\t<tr><td><span style=white-space:pre-wrap>greenkeeper.io &lt;support@greenkeeper.io&gt;           </span></td><td>1                                                                                               </td></tr>\n",
       "\t<tr><td><span style=white-space:pre-wrap>greenkeeper-keeper &lt;nodaguti+keeper@gmail.com&gt;    </span></td><td>1                                                                                               </td></tr>\n",
       "\t<tr><td>greenkeeper &lt;greenkeeper@users.noreply.github.com&gt;</td><td>0                                                       </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ll}\n",
       " auth & pcl\\\\\n",
       "\\hline\n",
       "\t dependabot{[}bot{]} <support@dependabot.com>           & 1                                                     \\\\\n",
       "\t greenkeeper <felixhennigfh@gmail.com>              & 0                                                 \\\\\n",
       "\t Greenkeeper <support@greenkeeper.io>               & 1                                                 \\\\\n",
       "\t greenkeeper.io <support@greenkeeper.io>            & 1                                                 \\\\\n",
       "\t greenkeeper-keeper <nodaguti+keeper@gmail.com>     & 1                                                 \\\\\n",
       "\t greenkeeper <greenkeeper@users.noreply.github.com> & 0                                                 \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "auth | pcl | \n",
       "|---|---|---|---|---|---|\n",
       "| dependabot[bot] <support@dependabot.com>           | 1                                                  | \n",
       "| greenkeeper <felixhennigfh@gmail.com>              | 0                                                  | \n",
       "| Greenkeeper <support@greenkeeper.io>               | 1                                                  | \n",
       "| greenkeeper.io <support@greenkeeper.io>            | 1                                                  | \n",
       "| greenkeeper-keeper <nodaguti+keeper@gmail.com>     | 1                                                  | \n",
       "| greenkeeper <greenkeeper@users.noreply.github.com> | 0                                                  | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  auth                                               pcl\n",
       "1 dependabot[bot] <support@dependabot.com>           1  \n",
       "2 greenkeeper <felixhennigfh@gmail.com>              0  \n",
       "3 Greenkeeper <support@greenkeeper.io>               1  \n",
       "4 greenkeeper.io <support@greenkeeper.io>            1  \n",
       "5 greenkeeper-keeper <nodaguti+keeper@gmail.com>     1  \n",
       "6 greenkeeper <greenkeeper@users.noreply.github.com> 0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Predicting Class\n",
    "pcl = predict(rfModel2, ftestdata, type='class')\n",
    "df.paper.c = data.frame(auth, pcl)\n",
    "head(df.paper.c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>auth</th><th scope=col>p</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td><span style=white-space:pre-wrap>dependabot[bot] &lt;support@dependabot.com&gt;          </span></td><td>0.93                                                                                            </td></tr>\n",
       "\t<tr><td><span style=white-space:pre-wrap>greenkeeper &lt;felixhennigfh@gmail.com&gt;             </span></td><td>0.42                                                                                            </td></tr>\n",
       "\t<tr><td><span style=white-space:pre-wrap>Greenkeeper &lt;support@greenkeeper.io&gt;              </span></td><td>0.85                                                                                            </td></tr>\n",
       "\t<tr><td><span style=white-space:pre-wrap>greenkeeper.io &lt;support@greenkeeper.io&gt;           </span></td><td>0.89                                                                                            </td></tr>\n",
       "\t<tr><td><span style=white-space:pre-wrap>greenkeeper-keeper &lt;nodaguti+keeper@gmail.com&gt;    </span></td><td>0.90                                                                                            </td></tr>\n",
       "\t<tr><td>greenkeeper &lt;greenkeeper@users.noreply.github.com&gt;</td><td>0.26                                                    </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ll}\n",
       " auth & p\\\\\n",
       "\\hline\n",
       "\t dependabot{[}bot{]} <support@dependabot.com>           & 0.93                                                  \\\\\n",
       "\t greenkeeper <felixhennigfh@gmail.com>              & 0.42                                              \\\\\n",
       "\t Greenkeeper <support@greenkeeper.io>               & 0.85                                              \\\\\n",
       "\t greenkeeper.io <support@greenkeeper.io>            & 0.89                                              \\\\\n",
       "\t greenkeeper-keeper <nodaguti+keeper@gmail.com>     & 0.90                                              \\\\\n",
       "\t greenkeeper <greenkeeper@users.noreply.github.com> & 0.26                                              \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "auth | p | \n",
       "|---|---|---|---|---|---|\n",
       "| dependabot[bot] <support@dependabot.com>           | 0.93                                               | \n",
       "| greenkeeper <felixhennigfh@gmail.com>              | 0.42                                               | \n",
       "| Greenkeeper <support@greenkeeper.io>               | 0.85                                               | \n",
       "| greenkeeper.io <support@greenkeeper.io>            | 0.89                                               | \n",
       "| greenkeeper-keeper <nodaguti+keeper@gmail.com>     | 0.90                                               | \n",
       "| greenkeeper <greenkeeper@users.noreply.github.com> | 0.26                                               | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  auth                                               p   \n",
       "1 dependabot[bot] <support@dependabot.com>           0.93\n",
       "2 greenkeeper <felixhennigfh@gmail.com>              0.42\n",
       "3 Greenkeeper <support@greenkeeper.io>               0.85\n",
       "4 greenkeeper.io <support@greenkeeper.io>            0.89\n",
       "5 greenkeeper-keeper <nodaguti+keeper@gmail.com>     0.90\n",
       "6 greenkeeper <greenkeeper@users.noreply.github.com> 0.26"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Predicting Probability\n",
    " p = predict(rfModel2, ftestdata, type='prob')[,2]\n",
    "df.paper.p = data.frame(auth, p)\n",
    "head(df.paper.p)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training the Ensemble Model (BIMAN) is similar:\n",
    "\n",
    "Generate data frame for test samples with their BICA probability (`p`), 1-BIM ratio (`ratio`), and name prediction from BIN (`name`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate bot data for BIMAN training - Either run this or use example data from the next cell\n",
    "bto = fread('/da4_data/play/botDetection/bot_template.out', header =F) # BIM o/p for bots\n",
    "colnames(bto) = c('auth','commits', 'templates', 'ratio')\n",
    "bot_data_all = merge(df.paper.p, bto) # df.paper.p: BICA probability for bots\n",
    "bot_data_all =  bot_data_all[,c(1,2,5)]\n",
    "bot_data_all$ratio = 1 - bot_data_all$ratio\n",
    "bot_data_all$name = 1 # BIN o/p for bots: set to 1 for demonstration\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "'data.frame':\t67 obs. of  5 variables:\n",
      " $ auth : Factor w/ 61 levels \"Alba bot <deploybot@nanoporetech.com>\",..: 1 2 4 5 6 7 8 9 32 33 ...\n",
      " $ p    : num  0.97 0.89 0.57 1 1 1 0.87 0.22 0.99 1 ...\n",
      " $ ratio: num  0 0 0.833 0 1 ...\n",
      " $ name : int  1 1 1 1 1 1 1 1 1 1 ...\n",
      " $ BOT  : int  1 1 1 1 1 1 1 1 1 1 ...\n"
     ]
    }
   ],
   "source": [
    "### THIS IS AN EXAMPLE CELL - DO NOT RUN IF YOU RAN THE CELL ABOVE\n",
    "# Showing pre-calculated data for example, similar structure for non bot data\n",
    "load('bot_data_all.Rdata')\n",
    "str(bot_data_all)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## You need to run similar steps for non-bots as well"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "set.seed(25)\n",
    "nb.aa = sample(13150,67)\n",
    "df.nb = nonbotdata[nb.aa,]\n",
    "nbaa = df.nb$Author\n",
    "df.nb$Author = NULL\n",
    "p = predict(BICA, df.nb, type='prob')[,2]\n",
    "df.nbp = data.frame(nbaa, p)\n",
    "colnames(df.nbp) = c('auth', 'p')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nbto = fread('/da4_data/play/botDetection/nonbot_template.out', header =F)\n",
    "colnames(nbto) = c('auth','commits', 'templates', 'ratio')\n",
    "\n",
    "df.nba = merge(df.nbp, nbto)\n",
    "df.nba =  df.nba[,c(1,2,5)]\n",
    "df.nba$ratio = 1 - df.nba$ratio\n",
    "df.nba$name = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Merge bot & non-bot data\n",
    "df.nba$BOT = 0\n",
    "df.b.all$BOT = 1\n",
    "df.f.all = data.frame(rbind(df.b.all, df.nba))\n",
    "df.f.all$BOT = factor(df.f.all$BOT)\n",
    "\n",
    "df.f.all$name = factor(df.f.all$name)\n",
    "ta.f = df.f.all$auth\n",
    "df.f.all$auth = NULL\n",
    "\n",
    "BOT = df.f.all$BOT\n",
    "df.f.all$BOT = NULL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Training the BIMAN model:\n",
    "set.seed(100)\n",
    "ind = 1:nrow(df.f.all)\n",
    "indTRAIN = sample(ind,round(0.8*length(ind)))\n",
    "indTEST <- ind[-indTRAIN]\n",
    "\n",
    "rfModel <- randomForest(x=data_all_biman[indTRAIN,],  y=BOT[indTRAIN], ntree=1000, importance=T) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Loading pre-trained BIMAN model - EITHER RUN THE TRAINING OR USE PRE_TRAINED DATA\n",
    "load('ensemble.Rdata')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0gAAANICAMAAADKOT/pAAADAFBMVEUAAAABAQECAgIDAwME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\nCAkQICRA4P8BE2C86WcRyCcAAAAASUVORK5CYII=",
      "text/plain": [
       "Plot with title \"Random Forest Model: Bot Detection\""
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "varImpPlot(rfModel, type=1, main= 'Random Forest Model: Bot Detection')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Name of the BICA model: BICA\n",
    "## Name of ensemble BIMAN model: rfModel"
   ]
  }
 ],
 "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}