BoW_analysis_cwe_capec.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:13.876891Z",
     "iopub.status.busy": "2022-05-29T23:43:13.873913Z",
     "iopub.status.idle": "2022-05-29T23:43:15.099607Z",
     "shell.execute_reply": "2022-05-29T23:43:15.098424Z"
    }
   },
   "outputs": [],
   "source": [
    "import json\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "from scikit_posthocs import posthoc_wilcoxon"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:15.107482Z",
     "iopub.status.busy": "2022-05-29T23:43:15.106140Z",
     "iopub.status.idle": "2022-05-29T23:43:15.124672Z",
     "shell.execute_reply": "2022-05-29T23:43:15.123589Z"
    }
   },
   "outputs": [],
   "source": [
    "f = open(\"results_100_with_BoW_cwe_capec_class_weight_5.json\")\n",
    "class_weight_5_results = json.load(f)\n",
    "\n",
    "f = open(\"results_100_with_BoW_cwe_capec.json\")\n",
    "results = json.load(f)\n",
    "\n",
    "f = open(\"results_100_with_BoW_cwe_capec_with_cve_2021.json\")\n",
    "results_with_cve = json.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:15.132764Z",
     "iopub.status.busy": "2022-05-29T23:43:15.131640Z",
     "iopub.status.idle": "2022-05-29T23:43:15.141072Z",
     "shell.execute_reply": "2022-05-29T23:43:15.142027Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.22922887261973057\n",
      "0.22437061577268605\n",
      "0.8020524017467249\n",
      "0.7934206695778748\n"
     ]
    }
   ],
   "source": [
    "BoW_fp = results[\"1\"][\"A0\"][\"fp\"]\n",
    "BoW_class_weight_5_fp = class_weight_5_results[\"1\"][\"A0\"][\"fp\"]\n",
    "\n",
    "BoW_acc = results[\"1\"][\"A0\"][\"acc\"]\n",
    "BoW_class_weight_5_acc = class_weight_5_results[\"1\"][\"A0\"][\"acc\"]\n",
    "\n",
    "print(np.mean(BoW_fp))\n",
    "print(np.mean(BoW_class_weight_5_fp))\n",
    "\n",
    "print(np.mean(BoW_acc))\n",
    "print(np.mean(BoW_class_weight_5_acc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:15.148196Z",
     "iopub.status.busy": "2022-05-29T23:43:15.147114Z",
     "iopub.status.idle": "2022-05-29T23:43:15.151657Z",
     "shell.execute_reply": "2022-05-29T23:43:15.150504Z"
    }
   },
   "outputs": [],
   "source": [
    "results = {\"1\": results[\"1\"], \"2\": results[\"2\"], \"3\": results[\"3\"]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:15.159947Z",
     "iopub.status.busy": "2022-05-29T23:43:15.158860Z",
     "iopub.status.idle": "2022-05-29T23:43:15.165344Z",
     "shell.execute_reply": "2022-05-29T23:43:15.164492Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('1', 'A0')\n",
      "0.8020524017467249\n"
     ]
    }
   ],
   "source": [
    "highest_val = 0\n",
    "highest_encoding = None\n",
    "metric = \"acc\"\n",
    "for encoding in results:\n",
    "    for data_combo in results[encoding]:\n",
    "        mean_metric = np.mean(results[encoding][data_combo][metric])\n",
    "        if mean_metric > highest_val:\n",
    "            highest_val = mean_metric\n",
    "            highest_encoding = (encoding, data_combo)\n",
    "\n",
    "print(highest_encoding)\n",
    "print(highest_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:15.176001Z",
     "iopub.status.busy": "2022-05-29T23:43:15.174901Z",
     "iopub.status.idle": "2022-05-29T23:43:15.451728Z",
     "shell.execute_reply": "2022-05-29T23:43:15.452505Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Encoding method:  2 , Data combo:  A1\n",
      "0.798617176128093\n",
      "Encoding method:  2 , Data combo:  A0\n",
      "0.8010189228529839\n",
      "Encoding method:  2 , Data combo:  A1 + MI\n",
      "0.782911208151383\n",
      "Encoding method:  2 , Data combo:  A1 + D\n",
      "0.776768558951965\n",
      "Encoding method:  2 , Data combo:  A1 + MI + D\n",
      "0.7763609898107715\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[Text(0, 0.5, 'Data combo'), Text(0.5, 0, 'Accuracy')]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "encoding = \"2\"\n",
    "\n",
    "data = {}\n",
    "for data_combo in results[encoding]:\n",
    "    print(\"Encoding method: \", encoding, \", Data combo: \", data_combo)\n",
    "    data[data_combo] = results[encoding][data_combo][\"acc\"]\n",
    "    print(np.mean(results[encoding][data_combo][\"acc\"]))\n",
    "\n",
    "data = pd.DataFrame(data)\n",
    "\n",
    "sns.boxplot(data=data, orient=\"h\").set(ylabel=\"Data combo\", xlabel=\"Accuracy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:15.461077Z",
     "iopub.status.busy": "2022-05-29T23:43:15.460157Z",
     "iopub.status.idle": "2022-05-29T23:43:15.674009Z",
     "shell.execute_reply": "2022-05-29T23:43:15.675677Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Encoding method:  1 , Data combo:  A0\n",
      "0.8020378457059679\n",
      "Encoding method:  2 , Data combo:  A0\n",
      "0.7998544395924309\n",
      "Encoding method:  3 , Data combo:  A0\n",
      "0.7998544395924309\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[Text(0, 0.5, 'Encoding'), Text(0.5, 0, 'Accuracy')]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "\n",
    "data_combo = \"A0\"\n",
    "data = {}\n",
    "for encoding in results:\n",
    "    print(\"Encoding method: \", encoding, \", Data combo: \", data_combo)\n",
    "    data[encoding] = results[encoding][data_combo][\"acc\"]\n",
    "    print(np.median(results[encoding][data_combo][\"acc\"]))\n",
    "\n",
    "data = pd.DataFrame(data)\n",
    "\n",
    "sns.boxplot(data=data, orient=\"h\").set(ylabel=\"Encoding\", xlabel=\"Accuracy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:15.707769Z",
     "iopub.status.busy": "2022-05-29T23:43:15.693449Z",
     "iopub.status.idle": "2022-05-29T23:43:15.723472Z",
     "shell.execute_reply": "2022-05-29T23:43:15.724326Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "A1 + MI + D\n",
      "1\n",
      "A1 + MI + D\n",
      "2\n",
      "A1 + MI + D\n",
      "3\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>4.734087e-13</td>\n",
       "      <td>6.394813e-17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.734087e-13</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.692324e-17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>6.394813e-17</td>\n",
       "      <td>1.692324e-17</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              1             2             3\n",
       "1  1.000000e+00  4.734087e-13  6.394813e-17\n",
       "2  4.734087e-13  1.000000e+00  1.692324e-17\n",
       "3  6.394813e-17  1.692324e-17  1.000000e+00"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_combo = \"A1 + MI + D\"\n",
    "accs = []\n",
    "encodings = []\n",
    "for encoding in results:\n",
    "    if data_combo != \"A0\" or encoding not in [\"2\", \"5\"]:\n",
    "        print(data_combo)\n",
    "        print(encoding)\n",
    "        encodings.append(encoding)\n",
    "        accs.append(results[encoding][data_combo][\"acc\"])\n",
    "\n",
    "rank_sums = posthoc_wilcoxon(accs, p_adjust=\"bonferroni\")\n",
    "rank_sums.columns = encodings\n",
    "rank_sums.index = encodings\n",
    "rank_sums"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:15.734719Z",
     "iopub.status.busy": "2022-05-29T23:43:15.733616Z",
     "iopub.status.idle": "2022-05-29T23:43:15.777291Z",
     "shell.execute_reply": "2022-05-29T23:43:15.776204Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "A0\n",
      "3\n",
      "A1\n",
      "3\n",
      "A1 + MI\n",
      "3\n",
      "A1 + D\n",
      "3\n",
      "A1 + MI + D\n",
      "3\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>A0</th>\n",
       "      <th>A1</th>\n",
       "      <th>A1 + MI</th>\n",
       "      <th>A1 + D</th>\n",
       "      <th>A1 + MI + D</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>A0</th>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>4.319367e-01</td>\n",
       "      <td>3.876605e-17</td>\n",
       "      <td>3.863284e-17</td>\n",
       "      <td>3.856313e-17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1</th>\n",
       "      <td>4.319367e-01</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>3.868956e-17</td>\n",
       "      <td>3.858925e-17</td>\n",
       "      <td>3.874856e-17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1 + MI</th>\n",
       "      <td>3.876605e-17</td>\n",
       "      <td>3.868956e-17</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>2.203408e-16</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1 + D</th>\n",
       "      <td>3.863284e-17</td>\n",
       "      <td>3.858925e-17</td>\n",
       "      <td>2.203408e-16</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>6.067520e-16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1 + MI + D</th>\n",
       "      <td>3.856313e-17</td>\n",
       "      <td>3.874856e-17</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>6.067520e-16</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       A0            A1       A1 + MI        A1 + D  \\\n",
       "A0           1.000000e+00  4.319367e-01  3.876605e-17  3.863284e-17   \n",
       "A1           4.319367e-01  1.000000e+00  3.868956e-17  3.858925e-17   \n",
       "A1 + MI      3.876605e-17  3.868956e-17  1.000000e+00  2.203408e-16   \n",
       "A1 + D       3.863284e-17  3.858925e-17  2.203408e-16  1.000000e+00   \n",
       "A1 + MI + D  3.856313e-17  3.874856e-17  1.000000e+00  6.067520e-16   \n",
       "\n",
       "              A1 + MI + D  \n",
       "A0           3.856313e-17  \n",
       "A1           3.874856e-17  \n",
       "A1 + MI      1.000000e+00  \n",
       "A1 + D       6.067520e-16  \n",
       "A1 + MI + D  1.000000e+00  "
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoding = \"3\"\n",
    "accs = []\n",
    "data_combo_list = []\n",
    "for data_combo in results[encoding]:\n",
    "    print(data_combo)\n",
    "    print(encoding)\n",
    "    data_combo_list.append(data_combo)\n",
    "    accs.append(results[encoding][data_combo][\"acc\"])\n",
    "rank_sums = posthoc_wilcoxon(accs, p_adjust=\"bonferroni\")\n",
    "rank_sums.columns = data_combo_list\n",
    "rank_sums.index = data_combo_list\n",
    "rank_sums"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:15.787300Z",
     "iopub.status.busy": "2022-05-29T23:43:15.786490Z",
     "iopub.status.idle": "2022-05-29T23:43:16.422020Z",
     "shell.execute_reply": "2022-05-29T23:43:16.420587Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Encoding method:  1 , Data combo:  A0\n",
      "Encoding method:  1 , Data combo:  A1\n",
      "Encoding method:  1 , Data combo:  A1 + MI\n",
      "Encoding method:  1 , Data combo:  A1 + D\n",
      "Encoding method:  1 , Data combo:  A1 + MI + D\n",
      "Encoding method:  2 , Data combo:  A1\n",
      "Encoding method:  2 , Data combo:  A0\n",
      "Encoding method:  2 , Data combo:  A1 + MI\n",
      "Encoding method:  2 , Data combo:  A1 + D\n",
      "Encoding method:  2 , Data combo:  A1 + MI + D\n",
      "Encoding method:  3 , Data combo:  A0\n",
      "Encoding method:  3 , Data combo:  A1\n",
      "Encoding method:  3 , Data combo:  A1 + MI\n",
      "Encoding method:  3 , Data combo:  A1 + D\n",
      "Encoding method:  3 , Data combo:  A1 + MI + D\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[Text(0, 0.5, '(Encoding, data combination)'),\n",
       " Text(0.5, 0, 'Accuracy'),\n",
       " Text(0.5, 1.0, 'BoW Experiments')]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data = {}\n",
    "for encoding in results:\n",
    "    for data_combo in results[encoding]:\n",
    "        print(\"Encoding method: \", encoding, \", Data combo: \", data_combo)\n",
    "        data[(encoding, data_combo)] = results[encoding][data_combo][\"acc\"]\n",
    "\n",
    "data = pd.DataFrame(data)\n",
    "\n",
    "sns.boxplot(data=data, orient=\"h\").set(\n",
    "    ylabel=\"(Encoding, data combination)\", xlabel=\"Accuracy\", title=\"BoW Experiments\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "98b0a9b7b4eaaa670588a142fd0a9b87eaafe866f1db4228be72b4211d12040f"
  },
  "kernelspec": {
   "display_name": "Python 3.7.10 64-bit ('base': conda)",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}