similarity_scores_boxplot.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Similarity scores computed and saved!\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import json\n",
    "import dateutil.parser as dparser\n",
    "\n",
    "# Levenshtein distance function\n",
    "def levenshtein_distance(s1, s2):\n",
    "    if len(s1) > len(s2):\n",
    "        s1, s2 = s2, s1\n",
    "\n",
    "    distances = range(len(s1) + 1)\n",
    "    for i2, c2 in enumerate(s2):\n",
    "        distances_ = [i2+1]\n",
    "        for i1, c1 in enumerate(s1):\n",
    "            if c1 == c2:\n",
    "                distances_.append(distances[i1])\n",
    "            else:\n",
    "                distances_.append(1 + min((distances[i1], distances[i1 + 1], distances_[-1])))\n",
    "        distances = distances_\n",
    "    return distances[-1]\n",
    "\n",
    "# Function to calculate similarity scores between two annotations\n",
    "def compute_similarity_scores(gpt_annotation_str, human_annotation_str):\n",
    "    if not isinstance(gpt_annotation_str, str) or not isinstance(human_annotation_str, str):\n",
    "        return {}\n",
    "\n",
    "    try:\n",
    "        gpt_annotation = json.loads(gpt_annotation_str)\n",
    "        human_annotation = json.loads(human_annotation_str)\n",
    "    except json.JSONDecodeError:\n",
    "        return {}\n",
    "\n",
    "    key_mapping = {}\n",
    "    for gpt_key in gpt_annotation:\n",
    "        mapped_key = min(human_annotation.keys(), key=lambda human_key: levenshtein_distance(gpt_key, human_key))\n",
    "        key_mapping[gpt_key] = mapped_key\n",
    "\n",
    "    scores = {}\n",
    "    for gpt_key, human_key in key_mapping.items():\n",
    "        gpt_value = gpt_annotation.get(gpt_key, \"\")\n",
    "        human_value = human_annotation.get(human_key, \"\")\n",
    "        \n",
    "        if isinstance(gpt_value, list):\n",
    "            gpt_value = ', '.join(map(str, gpt_value))\n",
    "        if isinstance(human_value, list):\n",
    "            human_value = ', '.join(map(str, human_value))\n",
    "        \n",
    "        if \"Date\" in gpt_key:\n",
    "            try:\n",
    "                gpt_date = dparser.parse(gpt_value, fuzzy=True).date()\n",
    "                human_date = dparser.parse(human_value, fuzzy=True).date()\n",
    "                scores[gpt_key] = 0 if gpt_date == human_date else 1\n",
    "            except:\n",
    "                scores[gpt_key] = levenshtein_distance(gpt_value, human_value)\n",
    "        else:\n",
    "            scores[gpt_key] = levenshtein_distance(gpt_value.lower(), human_value.lower())\n",
    "    \n",
    "    return scores\n",
    "\n",
    "# Load CSV file\n",
    "df = pd.read_csv('categories_atomic2.csv')\n",
    "\n",
    "# Compute similarity scores\n",
    "df_scores = df.apply(lambda row: compute_similarity_scores(row['gpt4_annotation'], row['human_annotation']), axis=1)\n",
    "df_scores_df = pd.DataFrame(df_scores.tolist())\n",
    "df_combined = pd.concat([df, df_scores_df], axis=1)\n",
    "\n",
    "# Save to a new CSV file\n",
    "df_combined.to_csv('categories_atomic2_with_scores2.csv', index=False)\n",
    "\n",
    "print(\"Similarity scores computed and saved!\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# Load the CSV file\n",
    "df_scores = pd.read_csv('categories_atomic2_with_scores.csv')\n",
    "\n",
    "# Extract columns corresponding to similarity scores\n",
    "score_columns = [col for col in df_scores.columns if col not in ['URL', 'annotated_text', 'gpt4_annotation', 'human_annotation']]\n",
    "\n",
    "# Set up the plot style\n",
    "sns.set(style=\"whitegrid\")\n",
    "\n",
    "# Draw the boxplots\n",
    "plt.figure(figsize=(15, 10))\n",
    "sns.boxplot(data=df_scores[score_columns])\n",
    "plt.xticks(rotation=45)\n",
    "plt.ylabel('Similarity Score')\n",
    "plt.title('Boxplots of Similarity Scores for Each Field')\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ai",
   "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": "3.11.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}