paraphrase_detector/notebooks/03_semantic_methods.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "07e239bf",
   "metadata": {},
   "source": [
    "Imports and setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "1638b7b97e3bd6f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-22T11:40:21.711998Z",
     "start_time": "2025-11-22T11:40:20.129376Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 29 test pairs\n"
     ]
    }
   ],
   "source": [
    "import spacy\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "from sentence_transformers import SentenceTransformer\n",
    "\n",
    "# Load models\n",
    "nlp = spacy.load(\"en_core_web_lg\")\n",
    "model_bert = SentenceTransformer(\"all-MiniLM-L6-v2\")\n",
    "\n",
    "test_pairs = [\n",
    "## STANDARD TEST PAIRS (from previous notebooks)\n",
    "    # Direct copies and near-copies\n",
    "    (\"The cat sat on the mat.\", \"The cat sat on the mat.\"),           # Exact copy\n",
    "    (\"The cat sat on the mat.\", \"The cat sat on the mat\"),            # No punctuation\n",
    "    (\"The cat sat on the mat.\", \"The  cat  sat  on  the  mat.\"),      # Extra spaces\n",
    "    \n",
    "    # Paraphrases with same meaning\n",
    "    (\"The cat sat on the mat.\", \"On the mat, the cat was sitting.\"),  # Structural change\n",
    "    (\"The cat sat on the mat.\", \"The feline rested on the rug.\"),     # Synonym replacement\n",
    "    (\"The quick brown fox jumps.\", \"A fast brown fox leaps.\"),        # Partial synonym\n",
    "    \n",
    "    # Different sentences\n",
    "    (\"The cat sat on the mat.\", \"The dog ran in the park.\"),          # Different content\n",
    "    (\"I love programming.\", \"She enjoys reading books.\"),             # Completely different\n",
    "    (\"The weather is nice today.\", \"It's raining outside.\"),          # Opposite meaning\n",
    "    \n",
    "    # Edge cases\n",
    "    (\"Short.\", \"Short.\"),                                             # Very short\n",
    "    (\"A B C D E F G\", \"A B C D E F G\"),                              # Repeated words\n",
    "    (\"\", \"\"),                                                         # Empty strings\n",
    "## ADDITIONAL TEST PAIRS (for semantic evaluation)\n",
    "    # Polysemy (word sense ambiguity) - Critical for BERT vs SpaCy\n",
    "    (\"He went to the bank to deposit money.\", \"He went to the river bank to fish.\"),           # Bank: financial vs. riverbank\n",
    "    (\"I saw the bat in the cave.\", \"The baseball bat is broken.\"),                             # Bat: animal vs. sports equipment\n",
    "    (\"The light is very bright.\", \"Can you carry this light package?\"),                        # Light: brightness vs. weight\n",
    "    \n",
    "    # Synonymy (different words, same meaning) - Tests semantic understanding\n",
    "    (\"The car is fast.\", \"The automobile is quick.\"),                                          # car/automobile, fast/quick\n",
    "    (\"He commenced the project.\", \"He started the work.\"),                                     # commenced/started, project/work\n",
    "    (\"The substantial building is ornate.\", \"The large building is decorated.\"),               # substantial/large, ornate/decorated\n",
    "    \n",
    "    # Negation (opposite meaning) - Tests if methods catch semantic inversion\n",
    "    (\"The weather is good.\", \"The weather is bad.\"),                                           # Direct negation\n",
    "    (\"The solution is simple.\", \"The problem is complex.\"),                                    # Opposite adjectives\n",
    "    (\"I like this movie.\", \"I dislike this movie.\"),                                           # Explicit negation\n",
    "    \n",
    "    # Semantic similarity without word overlap - Pure meaning test\n",
    "    (\"A man walks down the street.\", \"A person strolls along the road.\"),                      # Very few shared words, similar meaning\n",
    "    (\"The student studied the textbook.\", \"The pupil learned from the book.\"),                 # Different words, same semantic content\n",
    "    \n",
    "    # Partial overlap (high word similarity, low semantic similarity)\n",
    "    (\"The bank is on the river.\", \"The bank account is overdrawn.\"),                           # \"bank\" is only shared word; different meanings\n",
    "    (\"I read the book last night.\", \"The book was read by many people.\"),                      # Same key words (book, read) but different focus/meaning\n",
    "    \n",
    "    # Length variation (tests robustness to sentence length)\n",
    "    (\"Cat sat.\", \"The cat sat on the mat for hours and hours.\"),                               # Very different lengths, similar core meaning\n",
    "    (\"Go.\", \"You should go to the store and buy milk immediately.\"),                           # Minimal vs. detailed\n",
    "    \n",
    "    # Metaphor/Figurative language\n",
    "    (\"Time is money.\", \"Time has value.\"),                                                     # Metaphor vs. literal\n",
    "    (\"The world is a stage.\", \"Life is like a theater performance.\"),                          # Figurative expressions\n",
    "]\n",
    "\n",
    "print(f\"Loaded {len(test_pairs)} test pairs\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b79941bf4553fd6",
   "metadata": {},
   "source": [
    "## Method 1: SIF smooth inverse frequency\n",
    "Weighting word vectors based on frequency, emphasize information-dense words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "8a3c4314a90086fe",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-22T11:47:39.286432Z",
     "start_time": "2025-11-22T11:47:39.271377Z"
    }
   },
   "outputs": [],
   "source": [
    "def sentence_similarity_sif(sent1, sent2, a=0.001):\n",
    "    \"\"\"Compute similarity using SIF-weighted word vectors.\"\"\"\n",
    "    def get_weighted_vector(text):\n",
    "        doc = nlp(text)\n",
    "        vectors = []\n",
    "        weights = []\n",
    "        \n",
    "        for token in doc:\n",
    "            if token.has_vector and token.is_alpha and not token.is_stop:\n",
    "                prob = np.exp(token.prob) if token.prob > -19 else 1e-6\n",
    "                weight = a / (a + prob)\n",
    "                vectors.append(token.vector)\n",
    "                weights.append(weight)\n",
    "        \n",
    "        if not vectors:\n",
    "            return np.zeros(nlp.vocab.vectors_length)\n",
    "        \n",
    "        return np.average(vectors, axis=0, weights=weights)\n",
    "\n",
    "    v1 = get_weighted_vector(sent1)\n",
    "    v2 = get_weighted_vector(sent2)\n",
    "    \n",
    "    return cosine_similarity(v1.reshape(1, -1), v2.reshape(1, -1))[0][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "c51a0436",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SIF Similarity: 0.609\n"
     ]
    }
   ],
   "source": [
    "s1 = \"The cat ate the food.\"\n",
    "s2 = \"The feline devoured the sustenance.\"\n",
    "\n",
    "print(f\"SIF Similarity: {sentence_similarity_sif(s1, s2):.3f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f32b5695f554268",
   "metadata": {},
   "source": [
    "## Method 2: BERT - Transformer Embedding\n",
    "Use contectual embeddings to resolve polysemy issues"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "68a6757447e4a1c7",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-18T23:45:03.085563Z",
     "start_time": "2025-11-18T23:45:03.082190Z"
    }
   },
   "outputs": [],
   "source": [
    "def sentence_similarity_bert(sent1, sent2):\n",
    "    \"\"\"Compute similarity using SBERT contextual embeddings.\"\"\"\n",
    "    embeddings = model_bert.encode([sent1, sent2])\n",
    "    return cosine_similarity([embeddings[0]], [embeddings[1]])[0][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "420df430",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Bank (Money) vs Bank (River): 0.590\n",
      "Bank (Money) vs Institution:  0.641\n"
     ]
    }
   ],
   "source": [
    "sent_a = \"He went to the bank to deposit money.\"\n",
    "sent_b = \"He went to the river bank to fish.\"\n",
    "sent_c = \"He visited the financial institution.\"\n",
    "\n",
    "print(f\"Bank (Money) vs Bank (River): {sentence_similarity_bert(sent_a, sent_b):.3f}\")\n",
    "print(f\"Bank (Money) vs Institution:  {sentence_similarity_bert(sent_a, sent_c):.3f}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a9c3aa050f5bc0fe",
   "metadata": {},
   "source": [
    "### Semantic Evaluation\n",
    "Comparing simple average baseline, SIF and BERT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c100956f89d9b581",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Semantic Method Results:\n",
      "======================================================================\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>pair_id</th>\n",
       "      <th>Spacy Avg</th>\n",
       "      <th>Spacy SIF</th>\n",
       "      <th>BERT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0.984</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0.965</td>\n",
       "      <td>0.954</td>\n",
       "      <td>0.930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0.901</td>\n",
       "      <td>0.696</td>\n",
       "      <td>0.549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>0.942</td>\n",
       "      <td>0.930</td>\n",
       "      <td>0.945</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>0.857</td>\n",
       "      <td>0.642</td>\n",
       "      <td>0.067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>0.702</td>\n",
       "      <td>0.529</td>\n",
       "      <td>0.154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>0.845</td>\n",
       "      <td>0.603</td>\n",
       "      <td>0.468</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>10</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>11</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>12</td>\n",
       "      <td>0.908</td>\n",
       "      <td>0.707</td>\n",
       "      <td>0.590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>13</td>\n",
       "      <td>0.844</td>\n",
       "      <td>0.716</td>\n",
       "      <td>0.472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>14</td>\n",
       "      <td>0.803</td>\n",
       "      <td>0.694</td>\n",
       "      <td>0.310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>15</td>\n",
       "      <td>0.939</td>\n",
       "      <td>0.803</td>\n",
       "      <td>0.831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>16</td>\n",
       "      <td>0.909</td>\n",
       "      <td>0.619</td>\n",
       "      <td>0.778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>17</td>\n",
       "      <td>0.934</td>\n",
       "      <td>0.838</td>\n",
       "      <td>0.819</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>18</td>\n",
       "      <td>0.979</td>\n",
       "      <td>0.930</td>\n",
       "      <td>0.822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>19</td>\n",
       "      <td>0.885</td>\n",
       "      <td>0.623</td>\n",
       "      <td>0.620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>20</td>\n",
       "      <td>0.967</td>\n",
       "      <td>0.873</td>\n",
       "      <td>0.721</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>21</td>\n",
       "      <td>0.912</td>\n",
       "      <td>0.780</td>\n",
       "      <td>0.603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>22</td>\n",
       "      <td>0.874</td>\n",
       "      <td>0.712</td>\n",
       "      <td>0.652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>23</td>\n",
       "      <td>0.813</td>\n",
       "      <td>0.632</td>\n",
       "      <td>0.506</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>24</td>\n",
       "      <td>0.891</td>\n",
       "      <td>0.891</td>\n",
       "      <td>0.620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>25</td>\n",
       "      <td>0.761</td>\n",
       "      <td>0.755</td>\n",
       "      <td>0.628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>26</td>\n",
       "      <td>0.806</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>27</td>\n",
       "      <td>0.890</td>\n",
       "      <td>0.764</td>\n",
       "      <td>0.735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>28</td>\n",
       "      <td>0.897</td>\n",
       "      <td>0.721</td>\n",
       "      <td>0.442</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    pair_id  Spacy Avg  Spacy SIF   BERT\n",
       "0         0      1.000      1.000  1.000\n",
       "1         1      0.984      1.000  0.995\n",
       "2         2      1.000      1.000  1.000\n",
       "3         3      0.965      0.954  0.930\n",
       "4         4      0.901      0.696  0.549\n",
       "5         5      0.942      0.930  0.945\n",
       "6         6      0.857      0.642  0.067\n",
       "7         7      0.702      0.529  0.154\n",
       "8         8      0.845      0.603  0.468\n",
       "9         9      1.000      1.000  1.000\n",
       "10       10      1.000      1.000  1.000\n",
       "11       11      0.000      0.000  1.000\n",
       "12       12      0.908      0.707  0.590\n",
       "13       13      0.844      0.716  0.472\n",
       "14       14      0.803      0.694  0.310\n",
       "15       15      0.939      0.803  0.831\n",
       "16       16      0.909      0.619  0.778\n",
       "17       17      0.934      0.838  0.819\n",
       "18       18      0.979      0.930  0.822\n",
       "19       19      0.885      0.623  0.620\n",
       "20       20      0.967      0.873  0.721\n",
       "21       21      0.912      0.780  0.603\n",
       "22       22      0.874      0.712  0.652\n",
       "23       23      0.813      0.632  0.506\n",
       "24       24      0.891      0.891  0.620\n",
       "25       25      0.761      0.755  0.628\n",
       "26       26      0.806      0.000  0.202\n",
       "27       27      0.890      0.764  0.735\n",
       "28       28      0.897      0.721  0.442"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def sentence_similarity_avg(s1, s2):\n",
    "    d1, d2 = nlp(s1), nlp(s2)\n",
    "    v1 = np.mean([t.vector for t in d1 if t.has_vector], axis=0) if len(d1) > 0 else np.zeros(300)\n",
    "    v2 = np.mean([t.vector for t in d2 if t.has_vector], axis=0) if len(d2) > 0 else np.zeros(300)\n",
    "    return cosine_similarity([v1], [v2])[0][0]\n",
    "\n",
    "def evaluate_semantic_methods(pairs):\n",
    "    \"\"\"Evaluate semantic methods on test pairs without verbose text output.\"\"\"\n",
    "    methods = {\n",
    "        \"Spacy Avg\": sentence_similarity_avg,\n",
    "        \"Spacy SIF\": sentence_similarity_sif,\n",
    "        \"BERT\": sentence_similarity_bert\n",
    "    }\n",
    "\n",
    "    rows = []\n",
    "    for i, (sent1, sent2) in enumerate(pairs):\n",
    "        row = {\"pair_id\": i, \"sentence_1\": sent1, \"sentence_2\": sent2}\n",
    "        for name, func in methods.items():\n",
    "            score = func(sent1, sent2)\n",
    "            row[name] = score\n",
    "        rows.append(row)\n",
    "\n",
    "    return pd.DataFrame(rows)\n",
    "\n",
    "results_df = evaluate_semantic_methods(test_pairs)\n",
    "\n",
    "# Display only the scores (not sentences) for a cleaner comparison\n",
    "print(\"Semantic Method Results:\")\n",
    "print(\"=\" * 70)\n",
    "display(results_df[[\"pair_id\", \"Spacy Avg\", \"Spacy SIF\", \"BERT\"]].round(3))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e5192493",
   "metadata": {},
   "source": [
    "### Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "5e586947",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 5))\n",
    "melted = results_df.melt(\n",
    "    id_vars=[\"pair_id\"], \n",
    "    value_vars=[\"Spacy Avg\", \"Spacy SIF\", \"BERT\"], \n",
    "    var_name=\"Method\", \n",
    "    value_name=\"Similarity\"\n",
    ")\n",
    "\n",
    "sns.barplot(data=melted, x=\"pair_id\", y=\"Similarity\", hue=\"Method\")\n",
    "plt.title(\"Semantic Similarity Scores by Method\")\n",
    "plt.xlabel(\"Pair ID\")\n",
    "plt.ylabel(\"Similarity\")\n",
    "plt.ylim(0, 1)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d3b42bc4",
   "metadata": {},
   "source": [
    "Category Comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "57f4e206",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Performance by Category:\n",
      "================================================================================\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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">Spacy Avg</th>\n",
       "      <th colspan=\"2\" halign=\"left\">Spacy SIF</th>\n",
       "      <th colspan=\"2\" halign=\"left\">BERT</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Different</th>\n",
       "      <td>0.801</td>\n",
       "      <td>0.086</td>\n",
       "      <td>0.591</td>\n",
       "      <td>0.057</td>\n",
       "      <td>0.229</td>\n",
       "      <td>0.211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Edge Case</th>\n",
       "      <td>0.667</td>\n",
       "      <td>0.577</td>\n",
       "      <td>0.667</td>\n",
       "      <td>0.577</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Exact Copy</th>\n",
       "      <td>0.995</td>\n",
       "      <td>0.009</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.998</td>\n",
       "      <td>0.003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Figurative</th>\n",
       "      <td>0.893</td>\n",
       "      <td>0.005</td>\n",
       "      <td>0.742</td>\n",
       "      <td>0.031</td>\n",
       "      <td>0.588</td>\n",
       "      <td>0.208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Length Var</th>\n",
       "      <td>0.784</td>\n",
       "      <td>0.032</td>\n",
       "      <td>0.377</td>\n",
       "      <td>0.534</td>\n",
       "      <td>0.415</td>\n",
       "      <td>0.302</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Negation</th>\n",
       "      <td>0.944</td>\n",
       "      <td>0.051</td>\n",
       "      <td>0.809</td>\n",
       "      <td>0.163</td>\n",
       "      <td>0.721</td>\n",
       "      <td>0.101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Paraphrase</th>\n",
       "      <td>0.936</td>\n",
       "      <td>0.032</td>\n",
       "      <td>0.860</td>\n",
       "      <td>0.143</td>\n",
       "      <td>0.808</td>\n",
       "      <td>0.224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Partial Overlap</th>\n",
       "      <td>0.852</td>\n",
       "      <td>0.055</td>\n",
       "      <td>0.762</td>\n",
       "      <td>0.183</td>\n",
       "      <td>0.563</td>\n",
       "      <td>0.080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Polysemy</th>\n",
       "      <td>0.851</td>\n",
       "      <td>0.053</td>\n",
       "      <td>0.706</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.457</td>\n",
       "      <td>0.140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Semantic Sim</th>\n",
       "      <td>0.893</td>\n",
       "      <td>0.027</td>\n",
       "      <td>0.746</td>\n",
       "      <td>0.048</td>\n",
       "      <td>0.628</td>\n",
       "      <td>0.035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Synonymy</th>\n",
       "      <td>0.927</td>\n",
       "      <td>0.016</td>\n",
       "      <td>0.754</td>\n",
       "      <td>0.117</td>\n",
       "      <td>0.809</td>\n",
       "      <td>0.028</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Spacy Avg        Spacy SIF          BERT       \n",
       "                     mean    std      mean    std   mean    std\n",
       "category                                                       \n",
       "Different           0.801  0.086     0.591  0.057  0.229  0.211\n",
       "Edge Case           0.667  0.577     0.667  0.577  1.000  0.000\n",
       "Exact Copy          0.995  0.009     1.000  0.000  0.998  0.003\n",
       "Figurative          0.893  0.005     0.742  0.031  0.588  0.208\n",
       "Length Var          0.784  0.032     0.377  0.534  0.415  0.302\n",
       "Negation            0.944  0.051     0.809  0.163  0.721  0.101\n",
       "Paraphrase          0.936  0.032     0.860  0.143  0.808  0.224\n",
       "Partial Overlap     0.852  0.055     0.762  0.183  0.563  0.080\n",
       "Polysemy            0.851  0.053     0.706  0.011  0.457  0.140\n",
       "Semantic Sim        0.893  0.027     0.746  0.048  0.628  0.035\n",
       "Synonymy            0.927  0.016     0.754  0.117  0.809  0.028"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Manually assign categories to each pair\n",
    "categories = [\n",
    "    # Direct copies (0-2)\n",
    "    'Exact Copy', 'Exact Copy', 'Exact Copy',\n",
    "    # Paraphrases (3-5)\n",
    "    'Paraphrase', 'Paraphrase', 'Paraphrase',\n",
    "    # Different (6-8)\n",
    "    'Different', 'Different', 'Different',\n",
    "    # Edge cases (9-11)\n",
    "    'Edge Case', 'Edge Case', 'Edge Case',\n",
    "    # Polysemy (12-14)\n",
    "    'Polysemy', 'Polysemy', 'Polysemy',\n",
    "    # Synonymy (15-17)\n",
    "    'Synonymy', 'Synonymy', 'Synonymy',\n",
    "    # Negation (18-20)\n",
    "    'Negation', 'Negation', 'Negation',\n",
    "    # Semantic Similarity (21-22)\n",
    "    'Semantic Sim', 'Semantic Sim',\n",
    "    # Partial Overlap (23-24)\n",
    "    'Partial Overlap', 'Partial Overlap',\n",
    "    # Length Variation (25-26)\n",
    "    'Length Var', 'Length Var',\n",
    "    # Figurative (27-28)\n",
    "    'Figurative', 'Figurative'\n",
    "]\n",
    "\n",
    "results_df['category'] = categories\n",
    "\n",
    "# Group by category and show average scores\n",
    "print(\"\\nPerformance by Category:\")\n",
    "print(\"=\" * 80)\n",
    "category_stats = results_df.groupby('category')[[\"Spacy Avg\", \"Spacy SIF\", \"BERT\"]].agg(['mean', 'std']).round(3)\n",
    "display(category_stats)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf836250",
   "metadata": {},
   "source": [
    "Comparison Heatmap"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "66a4bb52",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(14, 8))\n",
    "\n",
    "# Create a pivot table for heatmap (pairs vs methods)\n",
    "heatmap_data = results_df[[\"pair_id\", \"Spacy Avg\", \"Spacy SIF\", \"BERT\"]].set_index(\"pair_id\")\n",
    "\n",
    "sns.heatmap(heatmap_data, annot=True, fmt='.2f', cmap='RdYlGn', cbar_kws={'label': 'Similarity'}, \n",
    "            vmin=0, vmax=1, linewidths=0.5)\n",
    "plt.title('Semantic Similarity Heatmap: All Pairs vs Methods', fontsize=14, fontweight='bold')\n",
    "plt.xlabel('Method')\n",
    "plt.ylabel('Pair ID')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "89686f33",
   "metadata": {},
   "source": [
    "A method with higher varience it may be less reliable while all 3 methods agreeing indicates that a pair is confident"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "8c0d18f9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Method Variance (higher = more disagreement):\n",
      "Spacy Avg    0.033659\n",
      "Spacy SIF    0.061698\n",
      "BERT         0.073351\n",
      "dtype: float64\n",
      "\n",
      "Method Std Dev:\n",
      "Spacy Avg    0.183464\n",
      "Spacy SIF    0.248391\n",
      "BERT         0.270834\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "\n",
    "method_variance = results_df[[\"Spacy Avg\", \"Spacy SIF\", \"BERT\"]].var()\n",
    "method_std = results_df[[\"Spacy Avg\", \"Spacy SIF\", \"BERT\"]].std()\n",
    "\n",
    "print(\"Method Variance (higher = more disagreement):\")\n",
    "print(method_variance)\n",
    "print(\"\\nMethod Std Dev:\")\n",
    "print(method_std)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a05b9608",
   "metadata": {},
   "source": [
    "Two methods may give similar absoloute scores when compared but rank individual pairs differently"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "9cf007fb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    pair_id  Spacy Avg_rank  Spacy SIF_rank  BERT_rank\n",
      "0         0            28.5            26.5       27.5\n",
      "1         1            25.0            26.5       24.0\n",
      "2         2            28.5            26.5       27.5\n",
      "3         3            22.0            24.0       22.0\n",
      "4         4            15.0            10.0        9.0\n",
      "5         5            21.0            22.0       23.0\n",
      "6         6             9.0             8.0        1.0\n",
      "7         7             2.0             3.0        2.0\n",
      "8         8             8.0             4.0        6.0\n",
      "9         9            27.0            29.0       27.5\n",
      "10       10            26.0            26.5       27.5\n",
      "11       11             1.0             1.5       25.0\n",
      "12       12            16.0            11.0       10.0\n",
      "13       13             7.0            13.0        7.0\n",
      "14       14             4.0             9.0        4.0\n",
      "15       15            20.0            18.0       21.0\n",
      "16       16            17.0             5.0       18.0\n",
      "17       17            19.0            19.0       19.0\n",
      "18       18            24.0            23.0       20.0\n",
      "19       19            11.0             6.0       13.0\n",
      "20       20            23.0            20.0       16.0\n",
      "21       21            18.0            17.0       11.0\n",
      "22       22            10.0            12.0       15.0\n",
      "23       23             6.0             7.0        8.0\n",
      "24       24            13.0            21.0       12.0\n",
      "25       25             3.0            15.0       14.0\n",
      "26       26             5.0             1.5        3.0\n",
      "27       27            12.0            16.0       17.0\n",
      "28       28            14.0            14.0        5.0\n"
     ]
    }
   ],
   "source": [
    "for method in [\"Spacy Avg\", \"Spacy SIF\", \"BERT\"]:\n",
    "    results_df[f'{method}_rank'] = results_df[method].rank()\n",
    "\n",
    "print(results_df[[\"pair_id\", \"Spacy Avg_rank\", \"Spacy SIF_rank\", \"BERT_rank\"]])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df98b471",
   "metadata": {},
   "source": [
    "Test methods against the polysemy problem"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "04f40f67",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Polysemy Test (Should prefer higher for synonyms, lower for different meanings):\n",
      "   pair_id               sentence_1  Spacy Avg  Spacy SIF      BERT\n",
      "6        6  The cat sat on the mat.   0.856538   0.641634  0.067119\n",
      "7        7      I love programming.   0.702400   0.528659  0.153708\n"
     ]
    }
   ],
   "source": [
    "polysemy_pairs = results_df.iloc[6:8]  # Rows 6-7 are the \"bank\" examples\n",
    "print(\"Polysemy Test (Should prefer higher for synonyms, lower for different meanings):\")\n",
    "print(polysemy_pairs[[\"pair_id\", \"sentence_1\", \"Spacy Avg\", \"Spacy SIF\", \"BERT\"]])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.13.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}