similarity_tactics.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fegxPFvil9yJ"
      },
      "source": [
        "\n",
        "\n",
        "1.   maybe try SciBERT or CyberBERT\n",
        "2.   try to compute the transformer without preprocessing bc. here we loose information about the structure of the sentence.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "JhAP2cDQoreg"
      },
      "outputs": [],
      "source": [
        "#approach is based on: BERT based embeddings (Transformers) + Cosine Similarity\n",
        "#inspired by: https://towardsdatascience.com/semantic-similarity-using-transformers-8f3cb5bf66d6\n",
        "#why i have choosen this approach: https://medium.com/@adriensieg/text-similarities-da019229c894\n",
        "\n",
        "# Install a pip package in the current Jupyter kernel\n",
        "import sys\n",
        "!{sys.executable} -m pip install transformers\n",
        "!{sys.executable} -m pip install sentence-transformers\n",
        "\n",
        "from sentence_transformers import SentenceTransformer, util\n",
        "import numpy as np"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "9Dr1sh_mo04P"
      },
      "outputs": [],
      "source": [
        "# List of models optimized for semantic textual similarity can be found at:\n",
        "# https://docs.google.com/spreadsheets/d/14QplCdTCDwEmTqrn1LH4yrbKvdogK4oQvYO1K1aPR5M/edit#gid=0\n",
        "model = SentenceTransformer('stsb-mpnet-base-v2')\n",
        "\n",
        "import sys\n",
        "!{sys.executable} -m pip install mitreattack-python\n",
        "\n",
        "#import the mitreattack-python library: https://mitreattack-python.readthedocs.io/en/latest/index.html\n",
        "#documentation for the API: https://mitreattack-python.readthedocs.io/en/latest/mitre_attack_data/mitre_attack_data.html#api-reference"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nJ4z8_89o8d8",
        "outputId": "c1764496-b464-43aa-a02f-496404603e32"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Retrieved 14 ATT&CK tactics.\n"
          ]
        }
      ],
      "source": [
        "from ctypes import sizeof\n",
        "from mitreattack.stix20 import MitreAttackData\n",
        "import json\n",
        "import csv\n",
        "\n",
        "#we use only the tactics for enterprise\n",
        "mitre_attack_data = MitreAttackData(\"/content/enterprise-attack.json\")\n",
        "\n",
        "#get all tactics\n",
        "tactics = mitre_attack_data.get_tactics(remove_revoked_deprecated=True) #set with or without subtechniques (True/False)\n",
        "print(f\"Retrieved {len(tactics)} ATT&CK tactics.\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "h9ds7aoNrE2n",
        "outputId": "2f187f70-a75d-40c7-d8fb-8e0109e425a8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Unzipping corpora/stopwords.zip.\n",
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Unzipping tokenizers/punkt.zip.\n",
            "[nltk_data] Downloading package wordnet to /root/nltk_data...\n",
            "[nltk_data] Downloading package omw-1.4 to /root/nltk_data...\n"
          ]
        }
      ],
      "source": [
        "from sentence_transformers import SentenceTransformer, util\n",
        "import csv\n",
        "import numpy as np\n",
        "import re\n",
        "import string\n",
        "\n",
        "#for stopwords\n",
        "import nltk\n",
        "from nltk.corpus import stopwords\n",
        "nltk.download('stopwords')\n",
        "nltk.download('punkt')\n",
        "nltk.download('wordnet')\n",
        "nltk.download('omw-1.4')\n",
        "from nltk.tokenize import word_tokenize\n",
        "from nltk.stem import WordNetLemmatizer"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "t6WySTE5heoQ"
      },
      "outputs": [],
      "source": [
        "# Initialize lemmatizer\n",
        "lemmatizer = WordNetLemmatizer()\n",
        "\n",
        "tactic_embeddings = {}\n",
        "for tactic in tactics:\n",
        "    technique_id = tactic.get('id')\n",
        "    technique_name = tactic.get('name')\n",
        "    technique_description = tactic.get('description')\n",
        "    # print(technique_name)\n",
        "    # print(technique_description)\n",
        "    # print()\n",
        "\n",
        "    # Regular expression to match text within parentheses and the parentheses themselves\n",
        "    parentheses_regex = re.compile(r'(\\(Citation|\\(http)[^)]*\\)')\n",
        "    # Regular expression to match the <code> and </code> statements\n",
        "    code_regex = re.compile(r'<code>|<\\/code>')\n",
        "    # Regular expression to remove the brackets, but not the content\n",
        "    brackets_regex = re.compile(r'[\\[\\]]')\n",
        "\n",
        "    # Remove links from the text\n",
        "    technique_description = re.sub(r'http\\S+', '', technique_description)\n",
        "\n",
        "    # Remove punctuation from the text\n",
        "    technique_description = technique_description.translate(str.maketrans('', '', string.punctuation))\n",
        "\n",
        "    # Convert the text to lowercase\n",
        "    technique_description = technique_description.lower()\n",
        "\n",
        "    # Remove parentheses, code tags, and brackets from the text\n",
        "    technique_description = parentheses_regex.sub('', technique_description)\n",
        "    technique_description = code_regex.sub('', technique_description)\n",
        "    technique_description = brackets_regex.sub('', technique_description)\n",
        "\n",
        "    # Tokenize text\n",
        "    text_tokens = word_tokenize(technique_description)\n",
        "\n",
        "    # Remove stopwords and lemmatize tokens\n",
        "    tokens_without_sw = [lemmatizer.lemmatize(word) for word in text_tokens if not word in stopwords.words()]\n",
        "\n",
        "    # Join tokens back into a filtered sentence\n",
        "    filtered_sentence = (\" \").join(tokens_without_sw)\n",
        "\n",
        "    # Update technique description\n",
        "    technique_description = filtered_sentence\n",
        "\n",
        "    embedding = model.encode(technique_description, convert_to_tensor=True, normalize_embeddings=True)\n",
        "    \n",
        "    tactic_embeddings[technique_name] = embedding\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "l22F1wvfsU1x"
      },
      "outputs": [],
      "source": [
        "#abstract\n",
        "#with open(r\"unique_talks_without_speakers.csv\", encoding='utf8') as csv_file:\n",
        "with open(r\"groundtruth.csv\", encoding='utf8') as csv_file:\n",
        "    csv_reader = csv.reader(csv_file, delimiter=';')\n",
        "    result = {}\n",
        "    count = 0\n",
        "    \n",
        "    for row in csv_reader:\n",
        "      #if \"advisen\".lower() in row[0].lower(): #only consider one specific conference\n",
        "      if True:\n",
        "        abstract = row[2]\n",
        "        if (len(abstract) < 20): #if smaller then the abstract is missing ... and then we don't count this entry\n",
        "          continue\n",
        "        #print(abstract)\n",
        "        \n",
        "        # Remove links from the text\n",
        "        abstract = re.sub(r'http\\S+', '', abstract)\n",
        "\n",
        "        # Remove punctuation from the text\n",
        "        abstract = abstract.translate(str.maketrans('', '', string.punctuation))\n",
        "\n",
        "        # Convert the text to lowercase\n",
        "        abstract = abstract.lower()\n",
        "\n",
        "        # Tokenize text\n",
        "        text_tokens = word_tokenize(abstract)\n",
        "\n",
        "        # Remove stopwords and lemmatize tokens\n",
        "        tokens_without_sw = [lemmatizer.lemmatize(word) for word in text_tokens if not word in stopwords.words()]\n",
        "\n",
        "        # Join tokens back into a filtered sentence\n",
        "        filtered_sentence = (\" \").join(tokens_without_sw)\n",
        "\n",
        "        # Update technique description\n",
        "        abstract = filtered_sentence\n",
        "        #print(abstract)\n",
        "        embedding1 = model.encode(abstract, convert_to_tensor=True, normalize_embeddings=True)\n",
        "\n",
        "        count += 1\n",
        "\n",
        "        for tactic in tactics:\n",
        "          tactic_name = tactic.get('name')\n",
        "          #print(f'Technique ID: {technique_id} - Name: {technique_name} - Description: {technique_description}')\n",
        "\n",
        "          # encode technique to get their embeddings\n",
        "          embedding2 = tactic_embeddings[tactic_name]\n",
        "\n",
        "          # compute similarity scores of two embeddings\n",
        "          cosine_score = util.pytorch_cos_sim(embedding1, embedding2)\n",
        "\n",
        "          tmp = result.get(tactic_name, 0)\n",
        "          result[tactic_name] = tmp + cosine_score.item()\n",
        "          #print(cosine_score, \" \", tactic_name)\n",
        "      print(count)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gh526QKUW89_"
      },
      "outputs": [],
      "source": [
        "for res in result.keys():\n",
        "  #print(res, result[res]/count)\n",
        "  tmp = result.get(res, 0)\n",
        "  result[res] = tmp/count"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ytfx0KhYnDLX",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "31afa679-4dfa-4cad-a2fe-746d8b677c52"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "                  Tactic  no classification\n",
            "0      Credential Access           0.358202\n",
            "1              Execution           0.434496\n",
            "2                 Impact           0.424810\n",
            "3            Persistence           0.362401\n",
            "4   Privilege Escalation           0.409221\n",
            "5       Lateral Movement           0.408227\n",
            "6        Defense Evasion           0.381770\n",
            "7           Exfiltration           0.412005\n",
            "8              Discovery           0.407813\n",
            "9             Collection           0.398080\n",
            "10  Resource Development           0.401235\n",
            "11        Reconnaissance           0.392776\n",
            "12   Command and Control           0.378096\n",
            "13        Initial Access           0.396072\n"
          ]
        }
      ],
      "source": [
        "from pandas.core.apply import relabel_result\n",
        "import pandas as pd\n",
        "\n",
        "df = pd.DataFrame({'Tactic': result.keys(),'no classification': result.values()})\n",
        "\n",
        "print(df)\n",
        "df.to_csv('no_classification.csv', header=True, index=False, encoding='utf-8')"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Load the CSV file into a pandas DataFrame\n",
        "df = pd.read_csv(\"hum_offensive (1).csv\", delimiter=',')\n",
        "\n",
        "# Set the first column as the DataFrame index\n",
        "df = df.set_index(df.columns[0])\n",
        "\n",
        "# Create a heatmap using the remaining columns as the data\n",
        "sns.heatmap(df.iloc[:, 0:], cmap=\"YlGnBu\")\n",
        "#sns.heatmap(df.iloc[:, 1:])\n",
        "\n",
        "plt.savefig('similarity.png', dpi=400, bbox_inches='tight')#change dpi for image resolution\n",
        "# Show the plot\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 527
        },
        "id": "-Hgsu1JmQ2f3",
        "outputId": "260423f0-b9c4-4024-fb88-61fb168df843"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}