link_probs_analysis_cwe_capec.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:54.241196Z",
     "iopub.status.busy": "2022-05-29T23:43:54.237795Z",
     "iopub.status.idle": "2022-05-29T23:43:54.688122Z",
     "shell.execute_reply": "2022-05-29T23:43:54.686920Z"
    }
   },
   "outputs": [],
   "source": [
    "import json\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:54.701860Z",
     "iopub.status.busy": "2022-05-29T23:43:54.700456Z",
     "iopub.status.idle": "2022-05-29T23:43:54.728645Z",
     "shell.execute_reply": "2022-05-29T23:43:54.727680Z"
    }
   },
   "outputs": [],
   "source": [
    "with open(\"w_dict.json\", \"r\") as f:\n",
    "    w_dict = json.load(f)\n",
    "with open(\"ap_dict.json\", \"r\") as f:\n",
    "    ap_dict = json.load(f)\n",
    "with open(\"technique_dict.json\", \"r\") as f:\n",
    "    technique_dict = json.load(f)\n",
    "with open(\"tactic_dict.json\", \"r\") as f:\n",
    "    tactic_dict = json.load(f)\n",
    "with open(\"cwe_names.json\", \"r\") as f:\n",
    "    cwe_names = json.load(f)\n",
    "with open(\"ap_names.json\", \"r\") as f:\n",
    "    ap_names = json.load(f)\n",
    "with open(\"technique_names.json\", \"r\") as f:\n",
    "    technique_names = json.load(f)\n",
    "with open(\"tactic_names.json\", \"r\") as f:\n",
    "    tactic_names = json.load(f)\n",
    "with open(\"ap_mitigation_descriptions.json\", \"r\") as f:\n",
    "    ap_mitigation_descriptions = json.load(f)\n",
    "with open(\"cwe_mitigation_descriptions.json\", \"r\") as f:\n",
    "    cwe_mitigation_descriptions = json.load(f)\n",
    "with open(\"tech_mitigation_names.json\", \"r\") as f:\n",
    "    tech_mitigation_names = json.load(f)\n",
    "with open(\"tech_detection_names.json\", \"r\") as f:\n",
    "    tech_detection_names = json.load(f)\n",
    "with open(\"ap_detection_descriptions.json\", \"r\") as f:\n",
    "    ap_detection_descriptions = json.load(f)\n",
    "with open(\"cwe_detection_descriptions.json\", \"r\") as f:\n",
    "    cwe_detection_descriptions = json.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:54.738065Z",
     "iopub.status.busy": "2022-05-29T23:43:54.736660Z",
     "iopub.status.idle": "2022-05-29T23:43:57.624580Z",
     "shell.execute_reply": "2022-05-29T23:43:57.625214Z"
    }
   },
   "outputs": [],
   "source": [
    "spaCy_probs = pd.read_csv(\"link_probs_spaCy_cwe_capec.csv\")\n",
    "BoW_probs = pd.read_csv(\"link_probs_BoW_spaCy_cwe_capec.csv\")\n",
    "BERT_probs = pd.read_csv(\"link_probs_BERT_cwe_capec.csv\")\n",
    "\n",
    "BoW_probs.drop(columns=\"spaCy link probability\", inplace=True)\n",
    "\n",
    "df = BoW_probs.merge(spaCy_probs, on=[\"CAPEC\", \"CWE\", \"CAPEC name\", \"CWE name\"])\n",
    "df = df.merge(BERT_probs, on=[\"CAPEC\", \"CWE\", \"CAPEC name\", \"CWE name\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:57.634027Z",
     "iopub.status.busy": "2022-05-29T23:43:57.632556Z",
     "iopub.status.idle": "2022-05-29T23:43:57.683094Z",
     "shell.execute_reply": "2022-05-29T23:43:57.682429Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/venv_BRON_ML/lib/python3.6/site-packages/ipykernel_launcher.py:3: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\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>CAPEC</th>\n",
       "      <th>CWE</th>\n",
       "      <th>CAPEC name</th>\n",
       "      <th>CWE name</th>\n",
       "      <th>BoW link probability</th>\n",
       "      <th>spaCy link probability</th>\n",
       "      <th>BERT link probability</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>239054</th>\n",
       "      <td>capec_00741</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Web Application Fingerprinting</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.96</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81096</th>\n",
       "      <td>capec_00665</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Forceful Browsing</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.94</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39778</th>\n",
       "      <td>capec_00619</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Manipulating Opaque Client-based Data Tokens</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.95</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>288379</th>\n",
       "      <td>capec_01011</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Use of Known Domain Credentials</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44982</th>\n",
       "      <td>capec_00663</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>AJAX Footprinting</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>216933</th>\n",
       "      <td>capec_00783</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>XML Routing Detour Attacks</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>138925</th>\n",
       "      <td>capec_00596</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Dictionary-based Password Attack</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>264989</th>\n",
       "      <td>capec_00922</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>HTTP Parameter Pollution (HPP)</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>373856</th>\n",
       "      <td>capec_00743</td>\n",
       "      <td>cwe_01441</td>\n",
       "      <td>Flash Parameter Injection</td>\n",
       "      <td>Improper Input Validation</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>404343</th>\n",
       "      <td>capec_00678</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Session Sidejacking</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.98</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>350669</th>\n",
       "      <td>capec_00652</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>User-Controlled Filename</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>369433</th>\n",
       "      <td>capec_00611</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Accessing/Intercepting/Modifying HTTP Cookies</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>365691</th>\n",
       "      <td>capec_00601</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Exploitation of Trusted Identifiers</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>421814</th>\n",
       "      <td>capec_00817</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>HTTP Verb Tampering</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4032</th>\n",
       "      <td>capec_00581</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Accessing Functionality Not Properly Constrain...</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>443848</th>\n",
       "      <td>capec_00924</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Cross-Domain Search Timing</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>420619</th>\n",
       "      <td>capec_00638</td>\n",
       "      <td>cwe_01911</td>\n",
       "      <td>Session Credential Falsification through Predi...</td>\n",
       "      <td>Incorrect Comparison</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>84517</th>\n",
       "      <td>capec_00645</td>\n",
       "      <td>cwe_01441</td>\n",
       "      <td>SQL Injection</td>\n",
       "      <td>Improper Input Validation</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>277193</th>\n",
       "      <td>capec_00925</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Padding Oracle Crypto Attack</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136769</th>\n",
       "      <td>capec_00657</td>\n",
       "      <td>cwe_01907</td>\n",
       "      <td>Using Escaped Slashes in Alternate Encoding</td>\n",
       "      <td>Protection Mechanism Failure</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>113634</th>\n",
       "      <td>capec_00795</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Hijacking a privileged process</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61396</th>\n",
       "      <td>capec_01099</td>\n",
       "      <td>cwe_01527</td>\n",
       "      <td>Use of Known Kerberos Credentials</td>\n",
       "      <td>Authentication Bypass by Spoofing</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>299111</th>\n",
       "      <td>capec_00631</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Poison Web Service Registry</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>276528</th>\n",
       "      <td>capec_00927</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Transparent Proxy Abuse</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>84549</th>\n",
       "      <td>capec_01055</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Cryptanalysis of Cellular Encryption</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>141472</th>\n",
       "      <td>capec_00671</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Adversary in the Middle (AiTM)</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>175465</th>\n",
       "      <td>capec_00650</td>\n",
       "      <td>cwe_01424</td>\n",
       "      <td>Using Unicode Encoding to Bypass Validation Logic</td>\n",
       "      <td>Incorrect Behavior Order: Validate Before Filter</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>231239</th>\n",
       "      <td>capec_00592</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Choosing Message Identifier</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>427262</th>\n",
       "      <td>capec_00595</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Command Delimiters</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>376140</th>\n",
       "      <td>capec_00609</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Leveraging Time-of-Check and Time-of-Use (TOCT...</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87632</th>\n",
       "      <td>capec_00610</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Hijacking a Privileged Thread of Execution</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146889</th>\n",
       "      <td>capec_00652</td>\n",
       "      <td>cwe_01925</td>\n",
       "      <td>User-Controlled Filename</td>\n",
       "      <td>Improper Neutralization of Special Elements in...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64584</th>\n",
       "      <td>capec_00609</td>\n",
       "      <td>cwe_01907</td>\n",
       "      <td>Leveraging Time-of-Check and Time-of-Use (TOCT...</td>\n",
       "      <td>Protection Mechanism Failure</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>384872</th>\n",
       "      <td>capec_01038</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>IP Address Blocking</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24650</th>\n",
       "      <td>capec_00639</td>\n",
       "      <td>cwe_01523</td>\n",
       "      <td>Reusing Session IDs (aka Session Replay)</td>\n",
       "      <td>Improper Authentication</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>301070</th>\n",
       "      <td>capec_00752</td>\n",
       "      <td>cwe_01441</td>\n",
       "      <td>IMAP/SMTP Command Injection</td>\n",
       "      <td>Improper Input Validation</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>431680</th>\n",
       "      <td>capec_00863</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>ICMP IP 'ID' Field Error Message Probe</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.94</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>470634</th>\n",
       "      <td>capec_00619</td>\n",
       "      <td>cwe_01527</td>\n",
       "      <td>Manipulating Opaque Client-based Data Tokens</td>\n",
       "      <td>Authentication Bypass by Spoofing</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>139226</th>\n",
       "      <td>capec_00635</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Rainbow Table Password Cracking</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>453878</th>\n",
       "      <td>capec_00619</td>\n",
       "      <td>cwe_01523</td>\n",
       "      <td>Manipulating Opaque Client-based Data Tokens</td>\n",
       "      <td>Improper Authentication</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>386098</th>\n",
       "      <td>capec_00659</td>\n",
       "      <td>cwe_01521</td>\n",
       "      <td>Using UTF-8 Encoding to Bypass Validation Logic</td>\n",
       "      <td>Improper Authorization</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>221625</th>\n",
       "      <td>capec_01107</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Adversary in the Browser (AiTB)</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>389710</th>\n",
       "      <td>capec_01068</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Analysis of Packet Timing and Sizes</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21748</th>\n",
       "      <td>capec_00595</td>\n",
       "      <td>cwe_01925</td>\n",
       "      <td>Command Delimiters</td>\n",
       "      <td>Improper Neutralization of Special Elements in...</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>490693</th>\n",
       "      <td>capec_00930</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Generic Cross-Browser Cross-Domain Theft</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47972</th>\n",
       "      <td>capec_00657</td>\n",
       "      <td>cwe_01521</td>\n",
       "      <td>Using Escaped Slashes in Alternate Encoding</td>\n",
       "      <td>Improper Authorization</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>150505</th>\n",
       "      <td>capec_00614</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>HTTP Response Splitting</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>202081</th>\n",
       "      <td>capec_00639</td>\n",
       "      <td>cwe_01907</td>\n",
       "      <td>Reusing Session IDs (aka Session Replay)</td>\n",
       "      <td>Protection Mechanism Failure</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>141299</th>\n",
       "      <td>capec_00749</td>\n",
       "      <td>cwe_01585</td>\n",
       "      <td>Exploiting Incorrectly Configured Access Contr...</td>\n",
       "      <td>Origin Validation Error</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>245255</th>\n",
       "      <td>capec_00970</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>SaaS User Request Forgery</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>245212</th>\n",
       "      <td>capec_01069</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Electromagnetic Side-Channel Attack</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauth...</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>243698</th>\n",
       "      <td>capec_00803</td>\n",
       "      <td>cwe_01441</td>\n",
       "      <td>Command Injection</td>\n",
       "      <td>Improper Input Validation</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.94</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              CAPEC        CWE  \\\n",
       "239054  capec_00741  cwe_01442   \n",
       "81096   capec_00665  cwe_01442   \n",
       "39778   capec_00619  cwe_01442   \n",
       "288379  capec_01011  cwe_01442   \n",
       "44982   capec_00663  cwe_01442   \n",
       "216933  capec_00783  cwe_01442   \n",
       "138925  capec_00596  cwe_01442   \n",
       "264989  capec_00922  cwe_01442   \n",
       "373856  capec_00743  cwe_01441   \n",
       "404343  capec_00678  cwe_01442   \n",
       "350669  capec_00652  cwe_01442   \n",
       "369433  capec_00611  cwe_01442   \n",
       "365691  capec_00601  cwe_01442   \n",
       "421814  capec_00817  cwe_01442   \n",
       "4032    capec_00581  cwe_01442   \n",
       "443848  capec_00924  cwe_01442   \n",
       "420619  capec_00638  cwe_01911   \n",
       "84517   capec_00645  cwe_01441   \n",
       "277193  capec_00925  cwe_01442   \n",
       "136769  capec_00657  cwe_01907   \n",
       "113634  capec_00795  cwe_01442   \n",
       "61396   capec_01099  cwe_01527   \n",
       "299111  capec_00631  cwe_01442   \n",
       "276528  capec_00927  cwe_01442   \n",
       "84549   capec_01055  cwe_01442   \n",
       "141472  capec_00671  cwe_01442   \n",
       "175465  capec_00650  cwe_01424   \n",
       "231239  capec_00592  cwe_01442   \n",
       "427262  capec_00595  cwe_01442   \n",
       "376140  capec_00609  cwe_01442   \n",
       "87632   capec_00610  cwe_01442   \n",
       "146889  capec_00652  cwe_01925   \n",
       "64584   capec_00609  cwe_01907   \n",
       "384872  capec_01038  cwe_01442   \n",
       "24650   capec_00639  cwe_01523   \n",
       "301070  capec_00752  cwe_01441   \n",
       "431680  capec_00863  cwe_01442   \n",
       "470634  capec_00619  cwe_01527   \n",
       "139226  capec_00635  cwe_01442   \n",
       "453878  capec_00619  cwe_01523   \n",
       "386098  capec_00659  cwe_01521   \n",
       "221625  capec_01107  cwe_01442   \n",
       "389710  capec_01068  cwe_01442   \n",
       "21748   capec_00595  cwe_01925   \n",
       "490693  capec_00930  cwe_01442   \n",
       "47972   capec_00657  cwe_01521   \n",
       "150505  capec_00614  cwe_01442   \n",
       "202081  capec_00639  cwe_01907   \n",
       "141299  capec_00749  cwe_01585   \n",
       "245255  capec_00970  cwe_01442   \n",
       "245212  capec_01069  cwe_01442   \n",
       "243698  capec_00803  cwe_01441   \n",
       "\n",
       "                                               CAPEC name  \\\n",
       "239054                     Web Application Fingerprinting   \n",
       "81096                                   Forceful Browsing   \n",
       "39778        Manipulating Opaque Client-based Data Tokens   \n",
       "288379                    Use of Known Domain Credentials   \n",
       "44982                                   AJAX Footprinting   \n",
       "216933                         XML Routing Detour Attacks   \n",
       "138925                   Dictionary-based Password Attack   \n",
       "264989                     HTTP Parameter Pollution (HPP)   \n",
       "373856                          Flash Parameter Injection   \n",
       "404343                                Session Sidejacking   \n",
       "350669                           User-Controlled Filename   \n",
       "369433      Accessing/Intercepting/Modifying HTTP Cookies   \n",
       "365691                Exploitation of Trusted Identifiers   \n",
       "421814                                HTTP Verb Tampering   \n",
       "4032    Accessing Functionality Not Properly Constrain...   \n",
       "443848                         Cross-Domain Search Timing   \n",
       "420619  Session Credential Falsification through Predi...   \n",
       "84517                                       SQL Injection   \n",
       "277193                       Padding Oracle Crypto Attack   \n",
       "136769        Using Escaped Slashes in Alternate Encoding   \n",
       "113634                     Hijacking a privileged process   \n",
       "61396                   Use of Known Kerberos Credentials   \n",
       "299111                        Poison Web Service Registry   \n",
       "276528                            Transparent Proxy Abuse   \n",
       "84549                Cryptanalysis of Cellular Encryption   \n",
       "141472                     Adversary in the Middle (AiTM)   \n",
       "175465  Using Unicode Encoding to Bypass Validation Logic   \n",
       "231239                        Choosing Message Identifier   \n",
       "427262                                 Command Delimiters   \n",
       "376140  Leveraging Time-of-Check and Time-of-Use (TOCT...   \n",
       "87632          Hijacking a Privileged Thread of Execution   \n",
       "146889                           User-Controlled Filename   \n",
       "64584   Leveraging Time-of-Check and Time-of-Use (TOCT...   \n",
       "384872                                IP Address Blocking   \n",
       "24650            Reusing Session IDs (aka Session Replay)   \n",
       "301070                        IMAP/SMTP Command Injection   \n",
       "431680             ICMP IP 'ID' Field Error Message Probe   \n",
       "470634       Manipulating Opaque Client-based Data Tokens   \n",
       "139226                    Rainbow Table Password Cracking   \n",
       "453878       Manipulating Opaque Client-based Data Tokens   \n",
       "386098    Using UTF-8 Encoding to Bypass Validation Logic   \n",
       "221625                    Adversary in the Browser (AiTB)   \n",
       "389710                Analysis of Packet Timing and Sizes   \n",
       "21748                                  Command Delimiters   \n",
       "490693           Generic Cross-Browser Cross-Domain Theft   \n",
       "47972         Using Escaped Slashes in Alternate Encoding   \n",
       "150505                            HTTP Response Splitting   \n",
       "202081           Reusing Session IDs (aka Session Replay)   \n",
       "141299  Exploiting Incorrectly Configured Access Contr...   \n",
       "245255                          SaaS User Request Forgery   \n",
       "245212                Electromagnetic Side-Channel Attack   \n",
       "243698                                  Command Injection   \n",
       "\n",
       "                                                 CWE name  \\\n",
       "239054  Exposure of Sensitive Information to an Unauth...   \n",
       "81096   Exposure of Sensitive Information to an Unauth...   \n",
       "39778   Exposure of Sensitive Information to an Unauth...   \n",
       "288379  Exposure of Sensitive Information to an Unauth...   \n",
       "44982   Exposure of Sensitive Information to an Unauth...   \n",
       "216933  Exposure of Sensitive Information to an Unauth...   \n",
       "138925  Exposure of Sensitive Information to an Unauth...   \n",
       "264989  Exposure of Sensitive Information to an Unauth...   \n",
       "373856                          Improper Input Validation   \n",
       "404343  Exposure of Sensitive Information to an Unauth...   \n",
       "350669  Exposure of Sensitive Information to an Unauth...   \n",
       "369433  Exposure of Sensitive Information to an Unauth...   \n",
       "365691  Exposure of Sensitive Information to an Unauth...   \n",
       "421814  Exposure of Sensitive Information to an Unauth...   \n",
       "4032    Exposure of Sensitive Information to an Unauth...   \n",
       "443848  Exposure of Sensitive Information to an Unauth...   \n",
       "420619                               Incorrect Comparison   \n",
       "84517                           Improper Input Validation   \n",
       "277193  Exposure of Sensitive Information to an Unauth...   \n",
       "136769                       Protection Mechanism Failure   \n",
       "113634  Exposure of Sensitive Information to an Unauth...   \n",
       "61396                   Authentication Bypass by Spoofing   \n",
       "299111  Exposure of Sensitive Information to an Unauth...   \n",
       "276528  Exposure of Sensitive Information to an Unauth...   \n",
       "84549   Exposure of Sensitive Information to an Unauth...   \n",
       "141472  Exposure of Sensitive Information to an Unauth...   \n",
       "175465   Incorrect Behavior Order: Validate Before Filter   \n",
       "231239  Exposure of Sensitive Information to an Unauth...   \n",
       "427262  Exposure of Sensitive Information to an Unauth...   \n",
       "376140  Exposure of Sensitive Information to an Unauth...   \n",
       "87632   Exposure of Sensitive Information to an Unauth...   \n",
       "146889  Improper Neutralization of Special Elements in...   \n",
       "64584                        Protection Mechanism Failure   \n",
       "384872  Exposure of Sensitive Information to an Unauth...   \n",
       "24650                             Improper Authentication   \n",
       "301070                          Improper Input Validation   \n",
       "431680  Exposure of Sensitive Information to an Unauth...   \n",
       "470634                  Authentication Bypass by Spoofing   \n",
       "139226  Exposure of Sensitive Information to an Unauth...   \n",
       "453878                            Improper Authentication   \n",
       "386098                             Improper Authorization   \n",
       "221625  Exposure of Sensitive Information to an Unauth...   \n",
       "389710  Exposure of Sensitive Information to an Unauth...   \n",
       "21748   Improper Neutralization of Special Elements in...   \n",
       "490693  Exposure of Sensitive Information to an Unauth...   \n",
       "47972                              Improper Authorization   \n",
       "150505  Exposure of Sensitive Information to an Unauth...   \n",
       "202081                       Protection Mechanism Failure   \n",
       "141299                            Origin Validation Error   \n",
       "245255  Exposure of Sensitive Information to an Unauth...   \n",
       "245212  Exposure of Sensitive Information to an Unauth...   \n",
       "243698                          Improper Input Validation   \n",
       "\n",
       "        BoW link probability  spaCy link probability  BERT link probability  \n",
       "239054                  0.98                    0.96                   1.00  \n",
       "81096                   0.98                    0.94                   1.00  \n",
       "39778                   0.95                    1.00                   0.99  \n",
       "288379                  0.95                    0.99                   0.99  \n",
       "44982                   1.00                    0.96                   0.99  \n",
       "216933                  0.97                    0.98                   0.98  \n",
       "138925                  0.95                    0.98                   0.98  \n",
       "264989                  0.96                    0.96                   0.98  \n",
       "373856                  0.96                    0.94                   0.98  \n",
       "404343                  0.98                    1.00                   0.97  \n",
       "350669                  0.97                    0.99                   0.97  \n",
       "369433                  0.97                    0.99                   0.97  \n",
       "365691                  0.94                    0.98                   0.97  \n",
       "421814                  0.95                    0.97                   0.97  \n",
       "4032                    0.94                    0.97                   0.97  \n",
       "443848                  0.97                    0.96                   0.97  \n",
       "420619                  0.96                    0.95                   0.97  \n",
       "84517                   0.98                    0.94                   0.97  \n",
       "277193                  0.96                    0.94                   0.97  \n",
       "136769                  0.95                    0.94                   0.97  \n",
       "113634                  0.98                    0.98                   0.96  \n",
       "61396                   0.94                    0.98                   0.96  \n",
       "299111                  0.97                    0.95                   0.96  \n",
       "276528                  0.97                    0.94                   0.96  \n",
       "84549                   0.96                    0.99                   0.95  \n",
       "141472                  0.95                    0.98                   0.95  \n",
       "175465                  1.00                    0.96                   0.95  \n",
       "231239                  1.00                    0.96                   0.95  \n",
       "427262                  0.98                    0.96                   0.95  \n",
       "376140                  0.97                    0.96                   0.95  \n",
       "87632                   0.94                    0.96                   0.95  \n",
       "146889                  0.97                    0.95                   0.95  \n",
       "64584                   0.98                    0.94                   0.95  \n",
       "384872                  0.97                    0.94                   0.95  \n",
       "24650                   0.94                    0.94                   0.95  \n",
       "301070                  0.94                    0.94                   0.95  \n",
       "431680                  0.94                    1.00                   0.94  \n",
       "470634                  0.94                    0.99                   0.94  \n",
       "139226                  0.98                    0.98                   0.94  \n",
       "453878                  0.95                    0.98                   0.94  \n",
       "386098                  1.00                    0.97                   0.94  \n",
       "221625                  0.95                    0.97                   0.94  \n",
       "389710                  0.95                    0.97                   0.94  \n",
       "21748                   1.00                    0.96                   0.94  \n",
       "490693                  0.97                    0.96                   0.94  \n",
       "47972                   0.96                    0.96                   0.94  \n",
       "150505                  0.96                    0.96                   0.94  \n",
       "202081                  0.94                    0.96                   0.94  \n",
       "141299                  0.97                    0.95                   0.94  \n",
       "245255                  0.97                    0.95                   0.94  \n",
       "245212                  0.95                    0.95                   0.94  \n",
       "243698                  0.94                    0.94                   0.94  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "threshold = 0.94\n",
    "top_link_probs = df[\n",
    "    (df[\"BoW link probability\"] >= threshold)\n",
    "    & (df[\"spaCy link probability\"] >= threshold)\n",
    "    & (df[\"BERT link probability\"] >= threshold)\n",
    "]\n",
    "top_link_probs.sort_values(\n",
    "    by=[\"BERT link probability\", \"spaCy link probability\", \"BoW link probability\"],\n",
    "    ascending=False,\n",
    "    inplace=True,\n",
    ")\n",
    "top_link_probs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:58.359190Z",
     "iopub.status.busy": "2022-05-29T23:43:58.326429Z",
     "iopub.status.idle": "2022-05-29T23:43:58.500846Z",
     "shell.execute_reply": "2022-05-29T23:43:58.499842Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_links_list = []\n",
    "\n",
    "for cwe in w_dict:\n",
    "    links_with_cwe = top_link_probs[top_link_probs[\"CWE\"] == cwe.split(\"/\")[1]]\n",
    "    max_index = min(len(links_with_cwe), 1)\n",
    "    links_with_cwe = links_with_cwe[:max_index]\n",
    "    top_links_list.append(links_with_cwe)\n",
    "\n",
    "top_links = pd.concat(top_links_list).drop_duplicates()\n",
    "len(top_links)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:58.514971Z",
     "iopub.status.busy": "2022-05-29T23:43:58.513416Z",
     "iopub.status.idle": "2022-05-29T23:43:58.536434Z",
     "shell.execute_reply": "2022-05-29T23:43:58.535516Z"
    }
   },
   "outputs": [],
   "source": [
    "f = open(\"cwe_temp.json\")\n",
    "cwe = json.load(f)\n",
    "\n",
    "f = open(\"capec_temp.json\")\n",
    "capec = json.load(f)\n",
    "\n",
    "top_links_dict = top_links.to_dict(\"list\")\n",
    "top_links_dict[\"CWE original ID\"] = []\n",
    "top_links_dict[\"CAPEC original ID\"] = []\n",
    "for cwe_id in top_links_dict[\"CWE\"]:\n",
    "    for w in cwe:\n",
    "        if cwe_id == w[\"_key\"]:\n",
    "            top_links_dict[\"CWE original ID\"].append(w[\"original_id\"])\n",
    "            break\n",
    "for ap_id in top_links_dict[\"CAPEC\"]:\n",
    "    for ap in capec:\n",
    "        if ap_id == ap[\"_key\"]:\n",
    "            top_links_dict[\"CAPEC original ID\"].append(ap[\"original_id\"])\n",
    "            break\n",
    "top_links = pd.DataFrame(top_links_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:58.546500Z",
     "iopub.status.busy": "2022-05-29T23:43:58.544028Z",
     "iopub.status.idle": "2022-05-29T23:43:58.564127Z",
     "shell.execute_reply": "2022-05-29T23:43:58.565137Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0888684903248416"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "link_probs_over_50 = df[\n",
    "    (df[\"BoW link probability\"] >= 0.5)\n",
    "    & (df[\"spaCy link probability\"] >= 0.5)\n",
    "    & (df[\"BERT link probability\"] >= 0.5)\n",
    "]\n",
    "len(link_probs_over_50) / len(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:58.594981Z",
     "iopub.status.busy": "2022-05-29T23:43:58.593781Z",
     "iopub.status.idle": "2022-05-29T23:43:58.600325Z",
     "shell.execute_reply": "2022-05-29T23:43:58.599247Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CAPEC</th>\n",
       "      <th>CWE</th>\n",
       "      <th>CAPEC name</th>\n",
       "      <th>CWE name</th>\n",
       "      <th>BoW link probability</th>\n",
       "      <th>spaCy link probability</th>\n",
       "      <th>BERT link probability</th>\n",
       "      <th>CWE original ID</th>\n",
       "      <th>CAPEC original ID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>capec_00650</td>\n",
       "      <td>cwe_01424</td>\n",
       "      <td>Using Unicode Encoding to Bypass Validation Logic</td>\n",
       "      <td>Incorrect Behavior Order: Validate Before Filter</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.95</td>\n",
       "      <td>181</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>capec_00743</td>\n",
       "      <td>cwe_01441</td>\n",
       "      <td>Flash Parameter Injection</td>\n",
       "      <td>Improper Input Validation</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.98</td>\n",
       "      <td>20</td>\n",
       "      <td>174</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>capec_00741</td>\n",
       "      <td>cwe_01442</td>\n",
       "      <td>Web Application Fingerprinting</td>\n",
       "      <td>Exposure of Sensitive Information to an Unauthorized Actor</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.96</td>\n",
       "      <td>1.00</td>\n",
       "      <td>200</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>capec_00659</td>\n",
       "      <td>cwe_01521</td>\n",
       "      <td>Using UTF-8 Encoding to Bypass Validation Logic</td>\n",
       "      <td>Improper Authorization</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.94</td>\n",
       "      <td>285</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>capec_00639</td>\n",
       "      <td>cwe_01523</td>\n",
       "      <td>Reusing Session IDs (aka Session Replay)</td>\n",
       "      <td>Improper Authentication</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.95</td>\n",
       "      <td>287</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>capec_01099</td>\n",
       "      <td>cwe_01527</td>\n",
       "      <td>Use of Known Kerberos Credentials</td>\n",
       "      <td>Authentication Bypass by Spoofing</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.96</td>\n",
       "      <td>290</td>\n",
       "      <td>652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>capec_00749</td>\n",
       "      <td>cwe_01585</td>\n",
       "      <td>Exploiting Incorrectly Configured Access Control Security Levels</td>\n",
       "      <td>Origin Validation Error</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.94</td>\n",
       "      <td>346</td>\n",
       "      <td>180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>capec_00657</td>\n",
       "      <td>cwe_01907</td>\n",
       "      <td>Using Escaped Slashes in Alternate Encoding</td>\n",
       "      <td>Protection Mechanism Failure</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.97</td>\n",
       "      <td>693</td>\n",
       "      <td>78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>capec_00638</td>\n",
       "      <td>cwe_01911</td>\n",
       "      <td>Session Credential Falsification through Prediction</td>\n",
       "      <td>Incorrect Comparison</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.97</td>\n",
       "      <td>697</td>\n",
       "      <td>59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>capec_00652</td>\n",
       "      <td>cwe_01925</td>\n",
       "      <td>User-Controlled Filename</td>\n",
       "      <td>Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection')</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>74</td>\n",
       "      <td>73</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         CAPEC        CWE  \\\n",
       "0  capec_00650  cwe_01424   \n",
       "1  capec_00743  cwe_01441   \n",
       "2  capec_00741  cwe_01442   \n",
       "3  capec_00659  cwe_01521   \n",
       "4  capec_00639  cwe_01523   \n",
       "5  capec_01099  cwe_01527   \n",
       "6  capec_00749  cwe_01585   \n",
       "7  capec_00657  cwe_01907   \n",
       "8  capec_00638  cwe_01911   \n",
       "9  capec_00652  cwe_01925   \n",
       "\n",
       "                                                         CAPEC name  \\\n",
       "0  Using Unicode Encoding to Bypass Validation Logic                  \n",
       "1  Flash Parameter Injection                                          \n",
       "2  Web Application Fingerprinting                                     \n",
       "3  Using UTF-8 Encoding to Bypass Validation Logic                    \n",
       "4  Reusing Session IDs (aka Session Replay)                           \n",
       "5  Use of Known Kerberos Credentials                                  \n",
       "6  Exploiting Incorrectly Configured Access Control Security Levels   \n",
       "7  Using Escaped Slashes in Alternate Encoding                        \n",
       "8  Session Credential Falsification through Prediction                \n",
       "9  User-Controlled Filename                                           \n",
       "\n",
       "                                                                                             CWE name  \\\n",
       "0  Incorrect Behavior Order: Validate Before Filter                                                     \n",
       "1  Improper Input Validation                                                                            \n",
       "2  Exposure of Sensitive Information to an Unauthorized Actor                                           \n",
       "3  Improper Authorization                                                                               \n",
       "4  Improper Authentication                                                                              \n",
       "5  Authentication Bypass by Spoofing                                                                    \n",
       "6  Origin Validation Error                                                                              \n",
       "7  Protection Mechanism Failure                                                                         \n",
       "8  Incorrect Comparison                                                                                 \n",
       "9  Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection')   \n",
       "\n",
       "   BoW link probability  spaCy link probability  BERT link probability  \\\n",
       "0  1.00                  0.96                    0.95                    \n",
       "1  0.96                  0.94                    0.98                    \n",
       "2  0.98                  0.96                    1.00                    \n",
       "3  1.00                  0.97                    0.94                    \n",
       "4  0.94                  0.94                    0.95                    \n",
       "5  0.94                  0.98                    0.96                    \n",
       "6  0.97                  0.95                    0.94                    \n",
       "7  0.95                  0.94                    0.97                    \n",
       "8  0.96                  0.95                    0.97                    \n",
       "9  0.97                  0.95                    0.95                    \n",
       "\n",
       "  CWE original ID CAPEC original ID  \n",
       "0  181             71                \n",
       "1  20              174               \n",
       "2  200             170               \n",
       "3  285             80                \n",
       "4  287             60                \n",
       "5  290             652               \n",
       "6  346             180               \n",
       "7  693             78                \n",
       "8  697             59                \n",
       "9  74              73                "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.set_option(\"display.max_colwidth\", 0)\n",
    "top_links"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:43:58.610387Z",
     "iopub.status.busy": "2022-05-29T23:43:58.609274Z",
     "iopub.status.idle": "2022-05-29T23:44:02.490254Z",
     "shell.execute_reply": "2022-05-29T23:44:02.489026Z"
    }
   },
   "outputs": [],
   "source": [
    "import spacy\n",
    "\n",
    "encode = spacy.load(\"en_core_web_lg\")\n",
    "\n",
    "names = list(top_links[\"CAPEC name\"]) + list(top_links[\"CWE name\"])\n",
    "\n",
    "names_embeddings = []\n",
    "\n",
    "for i in range(len(names)):\n",
    "    names_embeddings.append(encode(names[i]).vector)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:44:02.498853Z",
     "iopub.status.busy": "2022-05-29T23:44:02.498180Z",
     "iopub.status.idle": "2022-05-29T23:44:02.502555Z",
     "shell.execute_reply": "2022-05-29T23:44:02.501968Z"
    }
   },
   "outputs": [],
   "source": [
    "def get_hidden_state(model, text):\n",
    "    inputs = tokenizer(text.lower(), truncation=True, return_tensors=\"pt\")\n",
    "    outputs = model(**inputs, output_hidden_states=True)\n",
    "    hidden_states = outputs.hidden_states\n",
    "    return hidden_states[-1][:, 0, :].detach().cpu().numpy().flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:44:02.511334Z",
     "iopub.status.busy": "2022-05-29T23:44:02.510677Z",
     "iopub.status.idle": "2022-05-29T23:44:06.674760Z",
     "shell.execute_reply": "2022-05-29T23:44:06.674052Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertModel: ['cls.seq_relationship.bias', 'cls.predictions.bias', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.bias', 'cls.predictions.transform.dense.weight', 'cls.predictions.decoder.weight', 'cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.weight']\n",
      "- This IS expected if you are initializing BertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n"
     ]
    }
   ],
   "source": [
    "from transformers import AutoModel, AutoTokenizer, AutoModelForMaskedLM\n",
    "\n",
    "tokenizer = AutoTokenizer.from_pretrained(\"bert-base-uncased\")\n",
    "pretrained_model = AutoModel.from_pretrained(\"bert-base-uncased\")\n",
    "\n",
    "# model_path = \"bert_base\"\n",
    "# finetuned_model = AutoModelForMaskedLM.from_pretrained(model_path)\n",
    "\n",
    "names = list(top_links[\"CAPEC name\"]) + list(top_links[\"CWE name\"])\n",
    "\n",
    "names_embeddings = []\n",
    "\n",
    "for i in range(len(names)):\n",
    "    names_embeddings.append(get_hidden_state(pretrained_model, names[i]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:44:06.680936Z",
     "iopub.status.busy": "2022-05-29T23:44:06.680185Z",
     "iopub.status.idle": "2022-05-29T23:44:07.187939Z",
     "shell.execute_reply": "2022-05-29T23:44:07.188960Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "\n",
    "pca = PCA(n_components=2)\n",
    "data = pca.fit_transform(names_embeddings).T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-29T23:44:07.197798Z",
     "iopub.status.busy": "2022-05-29T23:44:07.196438Z",
     "iopub.status.idle": "2022-05-29T23:44:07.822204Z",
     "shell.execute_reply": "2022-05-29T23:44:07.823431Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x1080 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "sns.set_theme()\n",
    "\n",
    "x, y = data[0], data[1]\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(15, 15))\n",
    "ax.scatter(x, y)\n",
    "\n",
    "for i, term in enumerate(names):\n",
    "    ax.annotate(term, (x[i], y[i]), xytext=(3, 2), textcoords=\"offset points\")\n",
    "\n",
    "plt.title(\"BERT embeddings\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "fe5398c8201012d3ec3837fc46b21f6bb5cc60fe8bdaa4e264ad4d58fc1758ab"
  },
  "kernelspec": {
   "display_name": "Python 3.6.9 ('venv_BRON_ML')",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}