ruelbasedDA.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "df1c1659-1241-4b1c-a058-d99521a4551c",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import pandas as pd\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "643ef6fd-a1b2-4488-950a-cb66802455d5",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def fewRel(path):\n",
    "    data_list = []  # Create a list to collect dictionaries for each row\n",
    "    sentences = []\n",
    "    relations = []\n",
    "    entities = []\n",
    "    data = {}\n",
    "    for filename in os.listdir(path):\n",
    "        if not filename.endswith('.json'):\n",
    "            continue\n",
    "        fullname = os.path.join(path, filename)\n",
    "        \n",
    "        with open(fullname) as f:\n",
    "            fewRel_corpus = json.load(f)\n",
    "            \n",
    "            for relation, value in fewRel_corpus.items():\n",
    "                for row in value:\n",
    "                    sentences.append(row['tokens'])\n",
    "                    relations.append(relation)\n",
    "                    entities.append((row['h'][0].strip(),row['t'][0].strip()))\n",
    "        data['sentence'] = sentences\n",
    "        data['relation'] = relations\n",
    "        data['entities'] = entities\n",
    "    \n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "20ba5334-f8c6-4483-a649-aef25e6e0d4e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df = pd.DataFrame(fewRel('data/'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "abf2582e-8cb5-4230-885b-f4cb40134ae7",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def random_rows_per_category(group):\n",
    "    return group.sample(n=1, random_state=42) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "22b6a587-6c8e-4100-b341-057b2e05eaca",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "random_rows = df.groupby('relation', group_keys=False).apply(random_rows_per_category)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "0c5f5474-4aa2-4636-968a-c5848a79fc2e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "example = \"'The FBI has jurisdiction over federal and local government in Texas.' is augumented to 'he CCB has jurisdiction over regional and local government in Brazil.'\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "799ed2d2-2744-48a2-8951-d8746373f888",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "question = 'Write 5 augumented sentences of the following text' + example + ' text ='"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "2dd835ff-6625-4282-b4b5-79a2a6504d0f",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10/10 [00:00<00:00, 28.72it/s]\n"
     ]
    }
   ],
   "source": [
    "all_augumented_data = []\n",
    "for i in tqdm(range(10)):\n",
    "    random_rows = df.groupby('relation', group_keys=False).apply(random_rows_per_category)\n",
    "    for index, row in random_rows.iterrows():\n",
    "        single_aug_data = {}\n",
    "        sent = ' '.join(row['sentence'])\n",
    "        augumented_text = rule_based_augmentation(sent)\n",
    "        #print(process_data(augumented_text))\n",
    "        single_aug_data['sentences'] = augumented_text\n",
    "        single_aug_data['relation'] = row['relation']\n",
    "        single_aug_data['entities'] = row['entities']\n",
    "       # output = augumented_text +\", \" + row['relation'] + \", \" +  row['entities'][0] +\", \" +  row['entities'][1]\n",
    "        all_augumented_data.append(single_aug_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "f3130523-72fb-4a91-acc4-9d9512b2f2ba",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def save(file_path,data):\n",
    "    with open(file_path, \"w\") as json_file:\n",
    "        json.dump(data, json_file)\n",
    "\n",
    "    print(f'Data saved to {file_path}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "7590ed11-06fa-4d20-ad14-7029a6df4a0b",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data saved to augumented_data/fewrel_50_pr_rulebased_data.json\n"
     ]
    }
   ],
   "source": [
    "file_path = \"augumented_data/fewrel_50_pr_rulebased_data.json\"\n",
    "save(file_path,all_augumented_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "693e34e9-6c65-418d-86a6-6a9fe16f2182",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def nytFb(path):\n",
    "    dist = \"train\"\n",
    "    \n",
    "    temp_sent = \"\"\n",
    "    \n",
    "    temp_rel = \"\"\n",
    "    data = {}\n",
    "    data_list = []  # Create a list to collect dictionaries for each row\n",
    "    sentences = []\n",
    "    relations = []\n",
    "    entities = []\n",
    "   \n",
    "    for filename in os.listdir(path):\n",
    "        if not filename.endswith('.json'): continue\n",
    "        fullname = os.path.join(path, filename)\n",
    "       \n",
    "        if filename == \"test.json\":\n",
    "            dist = 'test'\n",
    "        elif filename == \"train.json\":\n",
    "            dist = 'train'\n",
    "        else:\n",
    "            dist = 'valid'\n",
    "        data2 = pd.read_json(r''+ fullname, lines=True)\n",
    "        \n",
    "        return data2\n",
    "       \n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "db842066-073a-4b8a-ad8f-ba412872b8ee",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_nyt = nytFb('data/nyt/')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "59ef43c9-515e-476c-a962-57108b798f4a",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def transform_row(row):\n",
    "    sentence = row['sentText']\n",
    "    relation = row['relationMentions'][0]['label'].split('/')[-1] \n",
    "    ent1 = row['entityMentions'][0]['text']\n",
    "    ent2 = row['entityMentions'][1]['text']\n",
    "    entities = (ent1, ent2)\n",
    "    return sentence, relation, entities\n",
    "\n",
    "# Apply the transformation function to each row and convert the result to a list of tuples\n",
    "transformed_data = data_nyt.apply(transform_row, axis=1)\n",
    "\n",
    "# Create a new DataFrame from the transformed data\n",
    "nyt_df = pd.DataFrame(transformed_data.tolist(), columns=['sentence', 'relation', 'entities'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "342d7dcf-bd3c-4782-b388-8a4af172780b",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "56196"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "eef856de-6fbe-4c0b-a77d-1f53b2450bc8",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10/10 [00:00<00:00, 15.12it/s]\n"
     ]
    }
   ],
   "source": [
    "all_augumented_data = []\n",
    "for i in tqdm(range(10)):\n",
    "    random_rows = nyt_df.groupby('relation', group_keys=False).apply(random_rows_per_category)\n",
    "    for index, row in random_rows.iterrows():\n",
    "        single_aug_data = {}\n",
    "        sent = row['sentence']\n",
    "        augumented_text = rule_based_augmentation(sent)\n",
    "        #print(process_data(augumented_text))\n",
    "        single_aug_data['sentences'] = augumented_text\n",
    "        single_aug_data['relation'] = row['relation']\n",
    "        single_aug_data['entities'] = row['entities']\n",
    "       # output = augumented_text +\", \" + row['relation'] + \", \" +  row['entities'][0] +\", \" +  row['entities'][1]\n",
    "        all_augumented_data.append(single_aug_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "5ba88326-1481-49c5-b686-938fe3f15ab8",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data saved to augumented_data/nyt_50_pr_rulebased_data.json\n"
     ]
    }
   ],
   "source": [
    "file_path = \"augumented_data/nyt_50_pr_rulebased_data.json\"\n",
    "save(file_path,all_augumented_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "96b098db-8a12-4d67-bb80-8db671bb71cb",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "relation\n",
      "contains                    30240\n",
      "nationality                  5219\n",
      "place_lived                  5024\n",
      "company                      3971\n",
      "capital                      2430\n",
      "neighborhood_of              2108\n",
      "place_of_birth               1580\n",
      "administrative_divisions     1518\n",
      "country                      1517\n",
      "place_of_death               1023\n",
      "children                      381\n",
      "founders                      377\n",
      "place_founded                 318\n",
      "major_shareholder_of           98\n",
      "teams                          88\n",
      "location                       86\n",
      "major_shareholders             84\n",
      "religion                       53\n",
      "advisors                       34\n",
      "geographic_distribution        26\n",
      "ethnicity                       9\n",
      "people                          9\n",
      "profession                      2\n",
      "industry                        1\n",
      "Name: count, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "category_counts = nyt_df['relation'].value_counts()\n",
    "\n",
    "print(category_counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "7f1e28e5-00de-4fd5-ac14-fd974c31d927",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "threshold = 100\n",
    "\n",
    "# Filter categories with counts less than the threshold\n",
    "categories_below_threshold = category_counts[category_counts < threshold]\n",
    "\n",
    "# Create two separate DataFrames based on the filter\n",
    "df_below_threshold = nyt_df[nyt_df['relation'].isin(categories_below_threshold.index)]\n",
    "df_above_threshold = nyt_df[~nyt_df['relation'].isin(categories_below_threshold.index)]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "b7a16a6f-7827-4a3d-865c-0a189e802eab",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing:   0%|          | 0/20 [00:00<?, ?it/s]\n",
      "11it [00:00, 552.19it/s]\n",
      "\n",
      "11it [00:00, 580.13it/s]\n",
      "\n",
      "11it [00:00, 817.89it/s]\n",
      "\n",
      "11it [00:00, 897.58it/s]\n",
      "\n",
      "11it [00:00, 1479.28it/s]\n",
      "Processing:  25%|██▌       | 5/20 [00:00<00:00, 46.33it/s]\n",
      "11it [00:00, 1466.59it/s]\n",
      "\n",
      "11it [00:00, 1491.33it/s]\n",
      "\n",
      "11it [00:00, 1484.90it/s]\n",
      "\n",
      "11it [00:00, 1497.67it/s]\n",
      "\n",
      "11it [00:00, 1499.18it/s]\n",
      "\n",
      "11it [00:00, 1489.07it/s]\n",
      "\n",
      "11it [00:00, 1498.94it/s]\n",
      "\n",
      "11it [00:00, 1546.11it/s]\n",
      "\n",
      "11it [00:00, 1481.13it/s]\n",
      "Processing:  70%|███████   | 14/20 [00:00<00:00, 70.22it/s]\n",
      "11it [00:00, 1492.63it/s]\n",
      "\n",
      "11it [00:00, 1493.55it/s]\n",
      "\n",
      "11it [00:00, 1499.72it/s]\n",
      "\n",
      "11it [00:00, 1493.21it/s]\n",
      "\n",
      "11it [00:00, 1485.19it/s]\n",
      "\n",
      "11it [00:00, 1498.31it/s]\n",
      "Processing: 100%|██████████| 20/20 [00:00<00:00, 71.81it/s]\n"
     ]
    }
   ],
   "source": [
    "all_augmented_data = []\n",
    "\n",
    "    # Iterate using tqdm for a progress bar\n",
    "for i in tqdm(range(20), desc=\"Processing\"):\n",
    "    random_rows = df_below_threshold.groupby('relation', group_keys=False).apply(random_rows_per_category)\n",
    "\n",
    "    for index, row in tqdm(random_rows.iterrows()):\n",
    "        single_aug_data = {}\n",
    "        sent = row['sentence']\n",
    "        augumented_text = rule_based_augmentation(sent)\n",
    "\n",
    "        single_aug_data['sentence'] = augumented_text\n",
    "        single_aug_data['relation'] = row['relation']\n",
    "        single_aug_data['entities'] = row['entities']\n",
    "\n",
    "        all_augmented_data.append(single_aug_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "21aef42c-feab-4ee3-b4e4-ee6a262810f6",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data saved to augumented_data/nyt_less100_100Plus_rulebased_data.json\n"
     ]
    }
   ],
   "source": [
    "file_path = \"augumented_data/nyt_less100_100Plus_rulebased_data.json\"\n",
    "save(file_path,all_augumented_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ffc2f206-4578-49c6-ba94-23ab82524c13",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d097e1f1-e8ac-4e50-9ab3-4457c4fa32bf",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dfa09e4f-2b81-49b7-aa9e-e1d59cc79091",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "32d0cae2-10c5-4a3a-9326-3d7443fe9f9e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c1ffdfb8-b591-4880-a375-b23fd6f30e0a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original Sentence:\n",
      "The workers often arrive in convoys , like the 47 repair vehicles from FirstEnergy , a utility holding company in Akron , Ohio , that pulled into the Coliseum parking lot Saturday.\n",
      "\n",
      "Augmented Sentences:\n",
      "The worker often arrive in convoys , like the 47 repair vehicles from FirstEnergy , a utility holding company in Akron , Ohio , that pulled into the Coliseum parking lot Saturday.\n",
      "The prole often arrive in convoys , like the 47 repair vehicles from FirstEnergy , a utility holding company in Akron , Ohio , that pulled into the Coliseum parking lot Saturday.\n",
      "The doer often arrive in convoys , like the 47 repair vehicles from FirstEnergy , a utility holding company in Akron , Ohio , that pulled into the Coliseum parking lot Saturday.\n",
      "The proletarian often arrive in convoys , like the 47 repair vehicles from FirstEnergy , a utility holding company in Akron , Ohio , that pulled into the Coliseum parking lot Saturday.\n",
      "The actor often arrive in convoys , like the 47 repair vehicles from FirstEnergy , a utility holding company in Akron , Ohio , that pulled into the Coliseum parking lot Saturday.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package wordnet to\n",
      "[nltk_data]     /upb/users/m/manzoor/profiles/unix/cs/nltk_data...\n",
      "[nltk_data]   Package wordnet is already up-to-date!\n"
     ]
    }
   ],
   "source": [
    "import nltk\n",
    "from nltk.corpus import wordnet\n",
    "\n",
    "nltk.download('wordnet')\n",
    "\n",
    "# Function to get synonyms of a word\n",
    "def get_synonyms(word):\n",
    "    synonyms = set()\n",
    "    for syn in wordnet.synsets(word):\n",
    "        for lemma in syn.lemmas():\n",
    "            synonyms.add(lemma.name())\n",
    "    return list(synonyms)\n",
    "\n",
    "# Function to perform rule-based data augmentation\n",
    "def rule_based_augmentation(sentence):\n",
    "    words = sentence.split()  # Tokenize the sentence\n",
    "    augmented_sentences = []\n",
    "\n",
    "    for i, word in enumerate(words):\n",
    "        synonyms = get_synonyms(word)\n",
    "        if synonyms:\n",
    "            for synonym in synonyms:\n",
    "                new_sentence = \" \".join(words[:i] + [synonym] + words[i+1:])\n",
    "                augmented_sentences.append(new_sentence)\n",
    "\n",
    "    return augmented_sentences[:5]\n",
    "\n",
    "# Example sentence for data augmentation\n",
    "input_sentence = \"The workers often arrive in convoys , like the 47 repair vehicles from FirstEnergy , a utility holding company in Akron , Ohio , that pulled into the Coliseum parking lot Saturday.\"\n",
    "\n",
    "# Perform rule-based data augmentation\n",
    "augmented_sentences = rule_based_augmentation(input_sentence)\n",
    "\n",
    "# Display the augmented sentences\n",
    "print(\"Original Sentence:\")\n",
    "print(input_sentence)\n",
    "\n",
    "print(\"\\nAugmented Sentences:\")\n",
    "for sentence in augmented_sentences:\n",
    "    print(sentence)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c1ac1470-a409-4c0b-8dd9-2cef292913a6",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Defaulting to user installation because normal site-packages is not writeable\n",
      "Collecting textstat\n",
      "  Downloading textstat-0.7.3-py3-none-any.whl (105 kB)\n",
      "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m105.1/105.1 kB\u001b[0m \u001b[31m4.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hCollecting pyphen (from textstat)\n",
      "  Downloading pyphen-0.14.0-py3-none-any.whl (2.0 MB)\n",
      "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.0/2.0 MB\u001b[0m \u001b[31m34.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
      "\u001b[?25hInstalling collected packages: pyphen, textstat\n",
      "Successfully installed pyphen-0.14.0 textstat-0.7.3\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install textstat\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "9cd1351c-6690-4212-ae15-edfe000a97a8",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Flesch-Kincaid Readability Score: 9.9\n"
     ]
    }
   ],
   "source": [
    "import textstat\n",
    "\n",
    "text = \"The ACIC has jurisdiction over federal and local government in Australia. \"\n",
    "\n",
    "# Calculate the Flesch-Kincaid score\n",
    "flesch_score = textstat.flesch_kincaid_grade(text)\n",
    "\n",
    "print(f\"Flesch-Kincaid Readability Score: {flesch_score}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "972cb3b7-fc7a-4834-b002-40e3f37b3646",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.metrics.pairwise import cosine_similarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "b90cd63e-cabf-4856-a6d2-29a657158ca3",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package punkt to\n",
      "[nltk_data]     /upb/users/m/manzoor/profiles/unix/cs/nltk_data...\n",
      "[nltk_data]   Unzipping tokenizers/punkt.zip.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nltk.download('punkt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "27deba52-394e-4b30-9663-9646968b9def",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "sentence1 = \"This is the first sentence for comparison.\"\n",
    "sentence2 = \"This is the second sentence for comparison.\"\n",
    "\n",
    "sentences = [sentence1, sentence2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "22769c6a-9eb3-4bf4-8531-fce1f660104a",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "vectorizer = CountVectorizer().fit_transform(sentences)\n",
    "vectors = vectorizer.toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "3c12b926-11e1-42f4-bee6-c570a7bda89e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "cosine_sim = cosine_similarity(vectors)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "7e7d5473-5d7d-48fa-9932-9fd9e8b754e2",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cosine Similarity Score: 0.8571428571428569\n"
     ]
    }
   ],
   "source": [
    "similarity_score = cosine_sim[0][1]\n",
    "print(f\"Cosine Similarity Score: {similarity_score}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "83f151fe-11d7-4bab-9496-128944acffdf",
   "metadata": {},
   "source": [
    "### Below is the code for last table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "f16a6904-3c4b-4cf4-a626-04d158049cc8",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def calculate_complexity(sentence):\n",
    "    try:\n",
    "        return textstat.flesch_kincaid_grade(sentence)\n",
    "    except:\n",
    "        return None  # Handle cases where the textstat library may throw an exception\n",
    "\n",
    "# Apply the function to the 'sentences' column\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "3cf4689b-d00e-4aa1-bd7b-5fd41a035a27",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea8cb116-d497-4d9e-b5bb-8fa746d39e3f",
   "metadata": {},
   "source": [
    "#### NYT data load"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "id": "90f23827-315e-4a80-a803-eb3e75bdbf33",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data = pd.read_csv('fewRel_entities.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "id": "981f2270-93a4-4f1e-9ab7-62808d7d4ed4",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_nyt = pd.read_csv('data/nyt.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "164404ce-3f1c-41b1-bd4f-bed7563dc5d5",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_nyt_falcon = pd.read_csv('data/nyt_aug_falcon_50.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "id": "a7b13e00-3acd-4dc4-862e-82a9e84231e8",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_nyt_llama = pd.read_csv('data/nyt_aug_llama_50.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "b34919d1-74be-446c-a524-9d5a3a0541c7",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_nyt_rule = pd.read_csv('data/nyt_aug_rule_50.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "id": "4984c800-0410-4167-a53f-b9eca840df9d",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_nyt_rule['complexity'] = data_nyt_rule['sentence'].apply(calculate_complexity)\n",
    "data_nyt_falcon['complexity'] = data_nyt_falcon['sentence'].apply(calculate_complexity)\n",
    "data_nyt_llama['complexity'] = data_nyt_llama['sentence'].apply(calculate_complexity)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "id": "1f957781-6174-4f1a-b60e-83d58803ca3f",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Complexity: 14.284166666666666, min complexity: 4.3 and max complexity: 24.4, standered deviation= 5.178031412932588\n",
      "Average Complexity: 11.1375, min complexity: 5.6 and max complexity: 21.3, standered deviation= 3.007997749120766\n",
      "Average Complexity: 12.904596622889304, min complexity: -15.7 and max complexity: 32.2, standered deviation= 5.172718927676214\n"
     ]
    }
   ],
   "source": [
    "details(data_nyt_rule)\n",
    "details(data_nyt_falcon)\n",
    "details(data_nyt_llama)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "id": "058d69cc-34b0-4e19-a14a-3671ac3f0fb5",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5.291666666666667\n",
      "2.158333333333333\n",
      "3.6397748592870545\n"
     ]
    }
   ],
   "source": [
    "print(numEntities(data_nyt_rule))\n",
    "print(numEntities(data_nyt_falcon))\n",
    "print(numEntities(data_nyt_llama))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "id": "b579cb2f-7eda-4a05-ad3c-3fcc62ec8af3",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Average number of tokens across all sentences: 42.791666666666664\n",
      "\n",
      "Average number of tokens across all sentences: 17.116666666666667\n",
      "\n",
      "Average number of tokens across all sentences: 36.15384615384615\n"
     ]
    }
   ],
   "source": [
    "claculateTokens(data_nyt_rule)\n",
    "claculateTokens(data_nyt_falcon)\n",
    "claculateTokens(data_nyt_llama)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "id": "0f2179e3-54a9-4572-9cdb-b2972d0ab286",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Average cosine similarity between sentences: 0.2456338983887971\n",
      "\n",
      "Average cosine similarity between sentences: 0.39441716727522363\n",
      "\n",
      "Average cosine similarity between sentences: 0.16971563518938332\n"
     ]
    }
   ],
   "source": [
    "\n",
    "avgCosineSim(data_nyt_rule)\n",
    "avgCosineSim(data_nyt_falcon)\n",
    "avgCosineSim(data_nyt_llama)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8193eee-ea89-4675-add1-afa9242e03e3",
   "metadata": {},
   "source": [
    "#### Fewrel dataload"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "id": "ff618d0c-bad8-4404-8c9e-4dc55f9458ad",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_fewrel_falcon = pd.read_csv('aug_falcon_entities.csv')\n",
    "data_fewrel_llama = pd.read_csv('aug_llama_entities_cleaned.csv')\n",
    "data_fewrel_rule = pd.read_csv('aug_rule_entities.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "id": "a438c6ba-71ae-4f6f-8db5-8c7ba21b179f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_fewrel_falcon['complexity'] = data_fewrel_falcon['sentence'].apply(calculate_complexity)\n",
    "data_fewrel_llama['complexity'] = data_fewrel_llama['sentence'].apply(calculate_complexity)\n",
    "data_fewrel_rule['complexity'] = data_fewrel_rule['sentence'].apply(calculate_complexity)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "id": "df01e9cd-5c06-4218-9b9c-c7825c9926d5",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Complexity: 11.62875, min complexity: 3.2 and max complexity: 20.6, standered deviation= 3.1047944149109363\n",
      "Average Complexity: 13.358605821056415, min complexity: 1.3 and max complexity: 50.9, standered deviation= 4.298414408086651\n",
      "Average Complexity: 10.678125, min complexity: 2.1 and max complexity: 20.9, standered deviation= 3.29674732183595\n"
     ]
    }
   ],
   "source": [
    "details(data_fewrel_falcon)\n",
    "details(data_fewrel_llama)\n",
    "details(data_fewrel_rule)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "id": "3c12c985-f0c7-4c22-be8d-0f41e62d5da3",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rule = 2.35\n",
      "llama = 3.943586058210564\n",
      "falcon = 4.109375\n"
     ]
    }
   ],
   "source": [
    "print(f'falcon = {numEntities(data_fewrel_falcon)}')\n",
    "print(f'llama = {numEntities(data_fewrel_llama)}')\n",
    "print(f'rule = {numEntities(data_fewrel_rule)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "id": "96a8d23e-9442-40e3-9c87-dcab1630a6c6",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Average number of tokens across all sentences: 24.875\n",
      "\n",
      "Average number of tokens across all sentences: 18.765625\n",
      "\n",
      "Average number of tokens across all sentences: 32.842256557671575\n"
     ]
    }
   ],
   "source": [
    "claculateTokens(data_fewrel_rule)\n",
    "claculateTokens(data_fewrel_falcon)\n",
    "claculateTokens(data_fewrel_llama)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "id": "0dba8b9b-8443-42d2-a33b-9db833e703e2",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Average cosine similarity between sentences: 0.1578490177926236\n",
      "\n",
      "Average cosine similarity between sentences: 0.3709857184793308\n",
      "\n",
      "Average cosine similarity between sentences: 0.1846967784163656\n"
     ]
    }
   ],
   "source": [
    "avgCosineSim(data_fewrel_rule)\n",
    "avgCosineSim(data_fewrel_falcon)\n",
    "avgCosineSim(data_fewrel_llama)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "id": "8503eb5e-e3cd-4d6c-9e25-780c9d91473d",
   "metadata": {
    "tags": []
   },
   "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>sentence</th>\n",
       "      <th>relation</th>\n",
       "      <th>entities</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Merpati flight 106 departed Jakarta ( CGK ) on...</td>\n",
       "      <td>P931</td>\n",
       "      <td>('tjq', 'tanjung pandan')</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>The name was at one point changed to Nottingha...</td>\n",
       "      <td>P931</td>\n",
       "      <td>('east midlands airport', 'nottingham')</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>It is a four - level stack interchange near Fo...</td>\n",
       "      <td>P931</td>\n",
       "      <td>('fort lauderdale-hollywood international airp...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>It is the main alternate of Jinnah Internation...</td>\n",
       "      <td>P931</td>\n",
       "      <td>('jinnah international airport', 'karachi')</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Nearby airports include Akwa Ibom Airport at O...</td>\n",
       "      <td>P931</td>\n",
       "      <td>('margaret ekpo international airport', 'calab...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            sentence relation  \\\n",
       "0  Merpati flight 106 departed Jakarta ( CGK ) on...     P931   \n",
       "1  The name was at one point changed to Nottingha...     P931   \n",
       "2  It is a four - level stack interchange near Fo...     P931   \n",
       "3  It is the main alternate of Jinnah Internation...     P931   \n",
       "4  Nearby airports include Akwa Ibom Airport at O...     P931   \n",
       "\n",
       "                                            entities  \n",
       "0                          ('tjq', 'tanjung pandan')  \n",
       "1            ('east midlands airport', 'nottingham')  \n",
       "2  ('fort lauderdale-hollywood international airp...  \n",
       "3        ('jinnah international airport', 'karachi')  \n",
       "4  ('margaret ekpo international airport', 'calab...  "
      ]
     },
     "execution_count": 195,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "id": "be215847-16ba-4ecb-bf5d-390dce2b9c5e",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Average number of tokens across all sentences: 25.00049107142857\n"
     ]
    }
   ],
   "source": [
    "claculateTokens(data)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "id": "1184694a-9211-4a83-8fd6-1ee08f49cf4c",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Average number of tokens across all sentences: 37.81715780482597\n"
     ]
    }
   ],
   "source": [
    "claculateTokens(data_nyt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "8ad283c4-6822-4e22-9ab3-a95bd97614da",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_nyt['complexity'] = data_nyt['sentence'].apply(calculate_complexity)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "60ccc1ec-532d-4822-aeb9-7a820526424e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data['complexity'] = data['sentence'].apply(calculate_complexity)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "474081c9-b05e-4a56-b94c-e670e2b2936c",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def details(data):    \n",
    "    average_complexity = data['complexity'].mean()\n",
    "    min_complexity = data['complexity'].min()\n",
    "    max_complexity = data['complexity'].max()\n",
    "    std_deviation_complexity = data['complexity'].std()\n",
    "\n",
    "    # Display the average complexity score\n",
    "    print(f'Average Complexity: {average_complexity}, min complexity: {min_complexity} and max complexity: {max_complexity}, standered deviation= {std_deviation_complexity}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "4caeba0a-a2b4-4d96-a3d4-b5a9b1d1886d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Complexity: 10.345794642857141, min complexity: -2.3 and max complexity: 25.6, standered deviation= 3.2103797695110985\n"
     ]
    }
   ],
   "source": [
    "details(data) ### Fewrel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "034c3c71-a805-45c0-8905-50dd935b67a3",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Complexity: 12.894590362303365, min complexity: -2.7 and max complexity: 40.0, standered deviation= 5.353198862675786\n"
     ]
    }
   ],
   "source": [
    "details(data_nyt) ### Nyt-fb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "id": "2d6b5a64-60e3-40b3-bf0f-884bdc6ed7d7",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Average cosine similarity between sentences: 0.1916743467349068\n"
     ]
    }
   ],
   "source": [
    "#avgCosineSim(data)\n",
    "avgCosineSim(data_nyt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "id": "242fcdda-84b4-41ec-b61a-c8d1552661cc",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Average cosine similarity between sentences: 0.1583082503517445\n"
     ]
    }
   ],
   "source": [
    "avgCosineSim(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "03c5f2e7-d99b-4977-b7fb-8a54864a520d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Assuming you have two DataFrames: df1 and df2 with 'complexity' column\n",
    "\n",
    "# Combine the datasets into a single DataFrame\n",
    "combined_df = pd.concat([data['complexity'], data_nyt['complexity']], axis=1)\n",
    "combined_df.columns = ['FewRel', 'NYT-FB']\n",
    "\n",
    "# Create a boxplot\n",
    "plt.figure(figsize=(10, 6))\n",
    "sns.boxplot(data=combined_df)\n",
    "plt.title('Comparison of Complexity Scores between FewRel and NYT-FB')\n",
    "plt.ylabel('Complexity Score')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "daf311e5-12a1-47c0-bccc-68029163825f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "nyt_aug_df = pd.concat([data_nyt['complexity'],data_nyt_falcon['complexity'], data_nyt_llama['complexity'], data_nyt_rule['complexity']], axis=1)\n",
    "nyt_aug_df.columns = ['NYT-FB','Falcon', 'Llama', 'Rule']\n",
    "\n",
    "# Create a boxplot\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "6835e3c8-7884-4439-b289-2264a471fb08",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "fewrel_aug_df= pd.concat([data['complexity'],data_fewrel_falcon['complexity'], data_fewrel_llama['complexity'], data_fewrel_rule['complexity']], axis=1)\n",
    "fewrel_aug_df.columns = ['FewRel','Falcon', 'Llama', 'Rule']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "id": "350f4e79-260d-40b9-8cd3-3a3ff693c58b",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 350x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Assuming you have a DataFrame nyt_aug_df with 'complexity' column\n",
    "\n",
    "plt.figure(figsize=(7/2, 5/2))\n",
    "\n",
    "ax = sns.boxplot(data=nyt_aug_df, showfliers=True, boxprops=dict(facecolor='None', edgecolor='green'), width=0.3, zorder=2)\n",
    "#ax = sns.boxplot(data=data_nyt_falcon['complexity'], showfliers=True, boxprops=dict(facecolor='None', edgecolor='green'), width=0.3, zorder=2,positions=[1])\n",
    "#ax = sns.boxplot(data=data_nyt_llama['complexity'], showfliers=True, boxprops=dict(facecolor='None', edgecolor='green'), width=0.3, zorder=2,positions=[1.5])\n",
    "#ax = sns.boxplot(data=data_nyt_rule['complexity'], showfliers=True, boxprops=dict(facecolor='None', edgecolor='green'), width=0.3, zorder=2,positions=[2.0])\n",
    "ax.set_aspect(0.05)\n",
    "\n",
    "\n",
    "sns.stripplot(data=nyt_aug_df, jitter=True, palette=\"Set2\", alpha=0.2, ax = ax,zorder=0)\n",
    "#sns.stripplot(data=data_nyt_falcon['complexity'], jitter=True, palette=\"Set3\", alpha=0.2, ax = ax,zorder=0)\n",
    "#sns.stripplot(data=data_nyt_llama['complexity'], jitter=True, palette=\"Set3\", alpha=0.2, ax = ax,zorder=0)\n",
    "#sns.stripplot(data=data_nyt_rule['complexity'], jitter=True, palette=\"Set3\", alpha=0.2, ax = ax,zorder=0)\n",
    "#plt.title('Comparison of Complexity Scores between Falcon, Llama, and Rule for NYT-FB dataset')\n",
    "plt.ylabel('Complexity Score')\n",
    "\n",
    "plt.tight_layout()  # Ensure the labels and title fit within the figure\n",
    "plt.savefig('images2/nyt_complexity.png', dpi=300)  # Save as PNG with 300 DPI\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "8661ac55-1e37-4d2b-b250-c0fa6768315b",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "relation_counts = data['relation'].value_counts()\n",
    "\n",
    "# Plot the bar graph\n",
    "relation_counts.plot(kind='bar')\n",
    "\n",
    "plt.title('Distribution of Relations')\n",
    "plt.xlabel('Relation')\n",
    "plt.ylabel('Count')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "a812a8d2-8b00-4f44-98cb-b13b0ebe554e",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data['relation_id'] = pd.factorize(data['relation'])[0]\n",
    "\n",
    "# Plot the histogram\n",
    "plt.hist(data['relation_id'], bins=len(data['relation'].unique()), align='left')\n",
    "\n",
    "plt.title('Distribution of Relations')\n",
    "plt.xlabel('Relation')\n",
    "plt.ylabel('Count')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "id": "bf44eaef-f9f7-4f51-9b23-83e3ab70bf8d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVsAAAD/CAYAAABfNXWhAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAjnklEQVR4nO3deVBUV/428KcVaAEBFYUWbbXdFyQKqKUYwQG3mDKWqUhGjdEgUVEQIaOQ/CK4BDQqGuOSEg24jEMyUTKSqOAWEsW4IIRFQlxwh8KFgAsBhfP+4eudaUHti92NtM+nqquGc0/f/nLKeubkcPoehRBCgIiIDKpRfRdARPQqYNgSERkBw5aIyAgYtkRERsCwJSIyAoYtEZERMGyJiIzArL4LeBlUV1fj+vXrsLGxgUKhqO9yiKiBEELgzp07cHJyQqNGz567MmwBXL9+HWq1ur7LIKIG6sqVK2jbtu0z+zBsAdjY2AB4NGC2trb1XA0RNRRlZWVQq9VShjwLwxaQlg5sbW0ZtkQkmy7Lj/wDGRGRETBsiYiMgGFLRGQEDFsiIiNg2BIRGQHDlojICLj1q46WZtys7xKIyADC+rY0yH05syUiMgKGLRGRETBsiYiMgGFLRGQEDFsiIiNg2BIRGQHDlojICBi2RERGwLAlIjIChi0RkREwbImIjIBhS0RkBAxbIiIjeOGwLSsrw/fff4+8vDx91ENEZJJkh+348eOxdu1aAEB5eTnc3d0xfvx4uLi4YOfOnXovkIjIFMgO259//hmvv/46ACAxMRFCCPz5559Ys2YNlixZovcCiYhMgeywLS0tRYsWLQAA+/btw9tvvw0rKyuMHj0aZ8+e1XuBRESmQHbYqtVqHDt2DPfu3cO+ffswfPhwAEBJSQmaNGmi9wKJiEyB7GNxgoODMXHiRDRt2hTt2rWDl5cXgEfLC71799Z3fUREJkF22AYEBKB///64cuUKhg0bhkaNHk2OO3bsyDVbIqKnqNOBj+7u7nBxcUFBQQE6deoEMzMzjB49Wt+1ERGZDNlrtvfv34efnx+srKzQq1cvXL58GQAQFBSEpUuX6r1AIiJTIDtsw8PD8dtvv+Gnn37S+oOYj48PvvnmG9kFXLt2DZMmTYK9vT2srKzQp08fpKenS9eFEIiMjISTkxMsLS3h5eWF3NxcrXtUVFQgMDAQLVu2hLW1NcaMGYOrV6/KroWIyFBkh+3333+PtWvXYvDgwVAoFFJ7z549cf78eVn3KikpgYeHB8zNzbF3716cOXMGK1euRLNmzaQ+n3/+OWJiYrB27VqcPHkSKpUKw4YNw507d6Q+wcHBSExMREJCAo4cOYK7d+/izTffRFVVldxfj4jIIGSv2d64cQMODg412u/du6cVvrpYtmwZ1Go14uLipLYOHTpI/1sIgdWrV+OTTz7BuHHjAABbtmyBo6MjduzYgenTp6O0tBSbN2/Gtm3b4OPjAwDYvn071Go1Dhw4gBEjRsj9FYmI9E72zLZfv3748ccfpZ8fB2xsbCwGDhwo6167d++Gu7s73nnnHTg4OKBv376IjY2VrhcUFKCoqEjaywsASqUSnp6eSEtLAwCkp6fjwYMHWn2cnJzg7Ows9XlSRUUFysrKtF5ERIYke2YbHR2NkSNH4syZM3j48CG++OIL5Obm4tixY0hNTZV1rwsXLmDDhg0ICQnBxx9/jBMnTiAoKAhKpRKTJ09GUVERAMDR0VHrfY6Ojrh06RIAoKioCBYWFmjevHmNPo/fX9vvsHDhQlm1EhG9CNkz20GDBuHo0aO4f/8+OnXqhJSUFDg6OuLYsWNwc3OTda/q6mq4uroiKioKffv2xfTp0+Hv748NGzZo9XtyeUII8dwli2f1CQ8PR2lpqfS6cuWKrLqJiOSq0z7b3r17Y8uWLS/84a1bt0bPnj212nr06CE9PUylUgF4NHtt3bq11Ke4uFia7apUKlRWVqKkpERrdltcXIxBgwbV+rlKpRJKpfKF6yci0pXsme2ePXuQnJxcoz05ORl79+6VdS8PDw/k5+drtf3xxx9o3749AECj0UClUmH//v3S9crKSqSmpkpB6ubmBnNzc60+hYWFyMnJeWrYEhEZm+ywDQsLq3VLlRACYWFhsu41d+5c/Prrr4iKisK5c+ewY8cObNy4EbNmzQLwaPkgODgYUVFRSExMRE5ODqZMmQIrKytMmDABAGBnZwc/Pz+Ehobi4MGDyMjIwKRJk9C7d29pdwIRUX2TvYxw9uzZGv/pDwDdu3fHuXPnZN2rX79+SExMRHh4OBYtWgSNRoPVq1dj4sSJUp958+ahvLwcAQEBKCkpwYABA5CSkgIbGxupz6pVq2BmZobx48ejvLwc3t7eiI+PR+PGjeX+ekREBqEQQgg5b1CpVNixYwf+9re/abUfOHAAEyZMQHFxsV4LNIaysjLY2dmhtLQUtra2Or1nacZNA1dFRPUhrG9LnfvKyQ7ZywhjxoxBcHCw1rfFzp07h9DQUIwZM0bu7YiIXgmyw3b58uWwtrZG9+7dodFooNFo0KNHD9jb22PFihWGqJGIqMGTvWZrZ2eHtLQ07N+/H7/99hssLS3h4uKCIUOGGKI+IiKTUKd9tgqFAsOHD9f6iiwRET1dncL24MGDOHjwIIqLi1FdXa117euvv9ZLYUREpkR22C5cuBCLFi2Cu7s7WrduLftJX0REryLZYfvVV18hPj4e7733niHqISIySbJ3I1RWVvJrsEREMskO22nTpmHHjh2GqIWIyGTJXkb466+/sHHjRhw4cAAuLi4wNzfXuh4TE6O34oiITIXssM3KykKfPn0AADk5OVrX+McyIqLayQ7bw4cPG6IOIiKTJnvN9rFz584hOTkZ5eXlAB49YpGIiGonO2xv3boFb29vdO3aFW+88QYKCwsBPPrDWWhoqN4LJCIyBbLDdu7cuTA3N8fly5dhZWUltfv6+mLfvn16LY6IyFTIXrNNSUlBcnIy2rZtq9XepUsX6cRbIiLSJntme+/ePa0Z7WM3b97kIYpERE8hO2yHDBmCrVu3Sj8rFApUV1dj+fLlGDp0qF6LIyIyFbKXEZYvXw4vLy+cOnUKlZWVmDdvHnJzc3H79m0cPXrUEDUSETV4sme2PXv2RFZWFvr3749hw4bh3r17GDduHDIyMtCpUydD1EhE1ODJntlevnwZarUaCxcurPVau3bt9FIYEZEpkT2z1Wg0uHHjRo32W7duQaPR6KUoIiJTIztshRC1PgPh7t27aNKkiV6KIiIyNTovI4SEhAB4tPvg008/1dr+VVVVhePHj0sPqCEiIm06h21GRgaARzPb7OxsWFhYSNcsLCzw2muv4aOPPtJ/hUREJkDnsH38tK+pU6fiiy++gK2trcGKIiIyNbJ3I8TFxRmiDiIikyY7bO/du4elS5c+9SjzCxcu6K04IiJTITtsp02bhtTUVLz33ns8ypyISEeyw3bv3r348ccf4eHhYYh6iIhMkux9ts2bN0eLFi0MUQsRkcmSHbaLFy/GggULcP/+fUPUQ0RkkmQvI6xcuRLnz5+Ho6MjOnToUOMo89OnT+utOCIiUyE7bMeOHWuAMoiITJvssI2IiDBEHUREJq1OR5n/+eef2LRpE8LDw3H79m0Aj5YPrl27ptfiiIhMheyZbVZWFnx8fGBnZ4eLFy/C398fLVq0QGJiIi5duqR1ZA4RET0ie2YbEhKCKVOm4OzZs1qPVBw1ahR+/vlnvRZHRGQqZIftyZMnMX369Brtbdq0QVFRkV6KIiIyNbLDtkmTJigrK6vRnp+fj1atWumlKCIiUyM7bN966y0sWrQIDx48APDoYeKXL19GWFgY3n77bb0XSERkCmSH7YoVK3Djxg04ODigvLwcnp6e6Ny5M2xsbPDZZ5/VuZDo6GgoFAoEBwdLbUIIREZGwsnJCZaWlvDy8kJubq7W+yoqKhAYGIiWLVvC2toaY8aMwdWrV+tcBxGRIcjejWBra4sjR47g0KFDOH36NKqrq+Hq6gofH586F3Hy5Els3LgRLi4uWu2ff/45YmJiEB8fj65du2LJkiUYNmwY8vPzYWNjAwAIDg5GUlISEhISYG9vj9DQULz55ptIT09H48aN61wTEZE+KYQQoj4LuHv3LlxdXbF+/XosWbIEffr0werVqyGEgJOTE4KDgzF//nwAj2axjo6OWLZsGaZPn47S0lK0atUK27Ztg6+vLwDg+vXrUKvV2LNnD0aMGKFTDWVlZbCzs0NpaanOJ1AszbhZt1+YiF5qYX1b6txXTnbovIxw/Phx7N27V6tt69at0Gg0cHBwwIcffoiKigqdi3xs1qxZGD16dI2ZcUFBAYqKijB8+HCpTalUwtPTE2lpaQCA9PR0PHjwQKuPk5MTnJ2dpT61qaioQFlZmdaLiMiQdA7byMhIZGVlST9nZ2fDz88PPj4+CAsLQ1JSEqKjo2V9eEJCAk6fPl3r+x5vI3N0dNRqd3R0lK4VFRXBwsICzZs3f2qf2kRHR8POzk56qdVqWXUTEcmlc9hmZmbC29tb+jkhIQEDBgxAbGwsQkJCsGbNGnz77bc6f/CVK1cwZ84cbN++XevLEU968iQIIcRzT4d4Xp/w8HCUlpZKrytXruhcNxFRXegctiUlJVqzzNTUVIwcOVL6uV+/frJCKz09HcXFxXBzc4OZmRnMzMyQmpqKNWvWwMzMTPqsJ2eoxcXF0jWVSoXKykqUlJQ8tU9tlEolbG1ttV5ERIakc9g6OjqioKAAAFBZWYnTp09j4MCB0vU7d+7UeLbts3h7eyM7OxuZmZnSy93dHRMnTkRmZiY6duwIlUqF/fv3S++prKxEamoqBg0aBABwc3ODubm5Vp/CwkLk5ORIfYiIXgY6b/0aOXIkwsLCsGzZMnz//fewsrLC66+/Ll3PyspCp06ddP5gGxsbODs7a7VZW1vD3t5eag8ODkZUVBS6dOmCLl26ICoqClZWVpgwYQIAwM7ODn5+fggNDYW9vT1atGiBjz76CL17936hrWhERPqmc9guWbIE48aNg6enJ5o2bYotW7bAwsJCuv71119r7QrQh3nz5qG8vBwBAQEoKSnBgAEDkJKSIu2xBYBVq1bBzMwM48ePR3l5Oby9vREfH889tkT0UpG9z7a0tBRNmzatEWa3b99G06ZNtQK4oeA+WyJ6zFD7bGV/g8zOzq7Wdp64S0T0dHU6qYGIiORh2BIRGQHDlojICHQKW1dXV+mLA4sWLcL9+/cNWhQRkanRKWzz8vJw7949AMDChQtx9+5dgxZFRGRqdNqN0KdPH0ydOhWDBw+GEAIrVqxA06ZNa+27YMECvRZIRGQKdArb+Ph4RERE4IcffoBCocDevXthZlbzrQqFgmFLRFQLncK2W7duSEhIAAA0atQIBw8ehIODg0ELIyIyJbK/1FBdXW2IOoiITJrssAWA8+fPY/Xq1cjLy4NCoUCPHj0wZ84cWQ+iISJ6lcjeZ5ucnIyePXvixIkTcHFxgbOzM44fP45evXppPeqQiIj+S/bMNiwsDHPnzsXSpUtrtM+fPx/Dhg3TW3FERKZC9sw2Ly8Pfn5+Ndo/+OADnDlzRi9FERGZGtlh26pVK2RmZtZoz8zM5A4FIqKnkL2M4O/vjw8//BAXLlzAoEGDoFAocOTIESxbtgyhoaGGqJGIqMGTHbaffvopbGxssHLlSoSHhwMAnJycEBkZiaCgIL0XSERkCmSHrUKhwNy5czF37lzcuXMHALSOqSEioprqtM/2MYYsEZFu+DxbIiIjYNgSERkBw5aIyAhkhe2DBw8wdOhQ/PHHH4aqh4jIJMkKW3Nzc+Tk5EChUBiqHiIikyR7GWHy5MnYvHmzIWohIjJZsrd+VVZWYtOmTdi/fz/c3d1hbW2tdT0mJkZvxRERmQrZYZuTkwNXV1cAqLF2y+UFIqLayQ7bw4cPG6IOIiKTVuetX+fOnUNycjLKy8sBAEIIvRVFRGRqZIftrVu34O3tja5du+KNN95AYWEhAGDatGl86hcR0VPIDtu5c+fC3Nwcly9fhpWVldTu6+uLffv26bU4IiJTIXvNNiUlBcnJyWjbtq1We5cuXXDp0iW9FUZEZEpkz2zv3bunNaN97ObNm1AqlXopiojI1MgO2yFDhmDr1q3SzwqFAtXV1Vi+fDmGDh2q1+KIiEyF7GWE5cuXw8vLC6dOnUJlZSXmzZuH3Nxc3L59G0ePHjVEjUREDZ7smW3Pnj2RlZWF/v37Y9iwYbh37x7GjRuHjIwMdOrUyRA1EhE1eHU6qUGlUmHhwoX6roWIyGTVKWxLSkqwefNm5OXlQaFQoEePHpg6dSpatGih7/qIiEyC7GWE1NRUaDQarFmzBiUlJbh9+zbWrFkDjUaD1NRUQ9RIRNTgyZ7Zzpo1C+PHj8eGDRvQuHFjAEBVVRUCAgIwa9Ys5OTk6L1IIqKGTvbM9vz58wgNDZWCFgAaN26MkJAQnD9/Xq/FERGZCtlh6+rqiry8vBrteXl56NOnjz5qIiIyOTqFbVZWlvQKCgrCnDlzsGLFChw5cgRHjhzBihUrMHfuXAQHB8v68OjoaPTr1w82NjZwcHDA2LFjkZ+fr9VHCIHIyEg4OTnB0tISXl5eyM3N1epTUVGBwMBAtGzZEtbW1hgzZgyuXr0qqxYiIkNSCB2ejdioUSMoFIrnPkZRoVCgqqpK5w8fOXIk3n33XfTr1w8PHz7EJ598guzsbJw5c0Y6AWLZsmX47LPPEB8fj65du2LJkiX4+eefkZ+fDxsbGwDAzJkzkZSUhPj4eNjb2yM0NBS3b99Genq61nLH05SVlcHOzg6lpaWwtbXVqfalGTd1/j2JqOEI69tS575yskOnsJXzgJn27dvr3PdJN27cgIODA1JTUzFkyBAIIeDk5ITg4GDMnz8fwKNZrKOjI5YtW4bp06ejtLQUrVq1wrZt2+Dr6wsAuH79OtRqNfbs2YMRI0bU+JyKigpUVFRIP5eVlUGtVjNsichgYavTboQXCVA5SktLAUDar1tQUICioiIMHz5c6qNUKuHp6Ym0tDRMnz4d6enpePDggVYfJycnODs7Iy0trdawjY6O5pcyiMio6vSlhmvXruHo0aMoLi5GdXW11rWgoKA6FSKEQEhICAYPHgxnZ2cAQFFREQDA0dFRq6+jo6M02y4qKoKFhQWaN29eo8/j9z8pPDwcISEh0s+PZ7ZERIYiO2zj4uIwY8YMWFhYwN7eXuuQR4VCUeewnT17NrKysnDkyJEa1548SFII8dzDJZ/VR6lU8nGQRGRUsrd+LViwAAsWLEBpaSkuXryIgoIC6XXhwoU6FREYGIjdu3fj8OHDWg8lV6lUAFBjhlpcXCzNdlUqFSorK1FSUvLUPkRE9U122N6/fx/vvvsuGjWq81mREiEEZs+ejV27duHQoUPQaDRa1zUaDVQqFfbv3y+1VVZWIjU1FYMGDQIAuLm5wdzcXKtPYWEhcnJypD5ERPVNdmL6+fnh3//+t14+fNasWdi+fTt27NgBGxsbFBUVoaioSDqxV6FQIDg4GFFRUUhMTEROTg6mTJkCKysrTJgwAQBgZ2cHPz8/hIaG4uDBg8jIyMCkSZPQu3dv+Pj46KVOIqIXJXvNNjo6Gm+++Sb27duH3r17w9zcXOt6TEyMzvfasGEDAMDLy0urPS4uDlOmTAEAzJs3D+Xl5QgICEBJSQkGDBiAlJQUaY8tAKxatQpmZmYYP348ysvL4e3tjfj4eJ322BIRGYNO+2z/1+LFixEREYFu3brB0dGxxh/IDh06pPciDY1faiCix+p1n+3/iomJwddffy3NPImI6Plkr9kqlUp4eHgYohYiIpMlO2znzJmDL7/80hC1EBGZLNnLCCdOnMChQ4fwww8/oFevXjX+QLZr1y69FUdEZCpkh22zZs0wbtw4Q9RCRGSy6vR1XSIikufFvwZGRETPJXtmq9FonvkQmLo+H4GIyJTJDtsnj7558OABMjIysG/fPvzjH//QV11ERCZFdtjOmTOn1vZ169bh1KlTL1wQEZEp0tua7ahRo7Bz50593Y6IyKToLWy/++476TgbIiLSJnsZoW/fvlp/IBNCoKioCDdu3MD69ev1WhwRkamQHbZjx47V+rlRo0Zo1aoVvLy80L17d33VRURkUmSHbUREhCHqICIyafxSAxGREeg8s23UqNFzT7RVKBR4+PDhCxdFRGRqdA7bxMTEp15LS0vDl19+CZmHPhARvTJ0Dtu33nqrRtvvv/+O8PBwJCUlYeLEiVi8eLFeiyMiMhV1WrO9fv06/P394eLigocPHyIzMxNbtmxBu3bt9F0fEZFJkBW2paWlmD9/Pjp37ozc3FwcPHgQSUlJcHZ2NlR9REQmQedlhM8//xzLli2DSqXCv/71r1qXFYiIqHY6H2XeqFEjWFpawsfHB40bN35qv4Z4LA6PMieix+r9KPPJkyc/d+sXERHVTuewjY+PN2AZRESmjd8gIyIyAoYtEZERMGyJiIyAYUtEZAQMWyIiI2DYEhEZAcOWiMgIGLZEREbAsCUiMgKGLRGRETBsiYiMgGFLRGQEDFsiIiNg2BIRGQHDlojICBi2RERGYDJhu379emg0GjRp0gRubm745Zdf6rskIiKJSYTtN998g+DgYHzyySfIyMjA66+/jlGjRuHy5cv1XRoREQATCduYmBj4+flh2rRp6NGjB1avXg21Wo0NGzbUd2lERABknEH2sqqsrER6ejrCwsK02ocPH460tLRa31NRUYGKigrp59LSUgCPTsrU1V9379ShWiJ62ZWVWcjo+ygzdDmkvMGH7c2bN1FVVQVHR0etdkdHRxQVFdX6nujoaCxcuLBGu1qtNkiNRNRw1EyG57tz5w7s7Oye2afBh+1jTx6zLoR46tHr4eHhCAkJkX6urq7G7du3YW9vL+u49rKyMqjValy5cuW5Z8bTf3Hc6o5jVzeGGjchBO7cuQMnJ6fn9m3wYduyZUs0bty4xiy2uLi4xmz3MaVSCaVSqdXWrFmzOtdga2vLf/h1wHGrO45d3Rhi3J43o32swf+BzMLCAm5ubti/f79W+/79+zFo0KB6qoqISFuDn9kCQEhICN577z24u7tj4MCB2LhxIy5fvowZM2bUd2lERABMJGx9fX1x69YtLFq0CIWFhXB2dsaePXvQvn17g36uUqlEREREjSUJejaOW91x7OrmZRg3hdBlzwIREb2QBr9mS0TUEDBsiYiMgGFLRGQEDFsiIiNg2L4APtbx2aKjo9GvXz/Y2NjAwcEBY8eORX5+vlYfIQQiIyPh5OQES0tLeHl5ITc3t54qfjlFR0dDoVAgODhYauO41e7atWuYNGkS7O3tYWVlhT59+iA9PV26Xq/jJqhOEhIShLm5uYiNjRVnzpwRc+bMEdbW1uLSpUv1XdpLY8SIESIuLk7k5OSIzMxMMXr0aNGuXTtx9+5dqc/SpUuFjY2N2Llzp8jOzha+vr6idevWoqysrB4rf3mcOHFCdOjQQbi4uIg5c+ZI7Ry3mm7fvi3at28vpkyZIo4fPy4KCgrEgQMHxLlz56Q+9TluDNs66t+/v5gxY4ZWW/fu3UVYWFg9VfTyKy4uFgBEamqqEEKI6upqoVKpxNKlS6U+f/31l7CzsxNfffVVfZX50rhz547o0qWL2L9/v/D09JTCluNWu/nz54vBgwc/9Xp9jxuXEerg8WMdhw8frtX+rMc60n8fZdmiRQsAQEFBAYqKirTGUalUwtPTk+MIYNasWRg9ejR8fHy02jlutdu9ezfc3d3xzjvvwMHBAX379kVsbKx0vb7HjWFbB3V5rOOrTgiBkJAQDB48GM7OzgAgjRXHsaaEhAScPn0a0dHRNa5x3Gp34cIFbNiwAV26dEFycjJmzJiBoKAgbN26FUD9j5tJfF23vsh5rOOrbvbs2cjKysKRI0dqXOM4arty5QrmzJmDlJQUNGnS5Kn9OG7aqqur4e7ujqioKABA3759kZubiw0bNmDy5MlSv/oaN85s66Auj3V8lQUGBmL37t04fPgw2rZtK7WrVCoA4Dg+IT09HcXFxXBzc4OZmRnMzMyQmpqKNWvWwMzMTBobjpu21q1bo2fPnlptPXr0kM4irO9/bwzbOuBjHXUjhMDs2bOxa9cuHDp0CBqNRuu6RqOBSqXSGsfKykqkpqa+0uPo7e2N7OxsZGZmSi93d3dMnDgRmZmZ6NixI8etFh4eHjW2Fv7xxx/SA6nq/d+bwf8EZ6Ieb/3avHmzOHPmjAgODhbW1tbi4sWL9V3aS2PmzJnCzs5O/PTTT6KwsFB63b9/X+qzdOlSYWdnJ3bt2iWys7PF3//+91d+C1Nt/nc3ghAct9qcOHFCmJmZic8++0ycPXtW/POf/xRWVlZi+/btUp/6HDeG7QtYt26daN++vbCwsBCurq7SliZ6BECtr7i4OKlPdXW1iIiIECqVSiiVSjFkyBCRnZ1df0W/pJ4MW45b7ZKSkoSzs7NQKpWie/fuYuPGjVrX63Pc+IhFIiIj4JotEZERMGyJiIyAYUtEZAQMWyIiI2DYEhEZAcOWiMgIGLZEREbAsCUiMgKGLb0yLl68CIVCgczMzJfiPvRqYdhSgzFlyhQoFAooFAqYmZmhXbt2mDlzJkpKSgz6mWPHjtVqU6vVKCwslJ7LS6QLhi01KCNHjkRhYSEuXryITZs2ISkpCQEBAUatoXHjxlCpVDAz4+OgSXcMW2pQlEolVCoV2rZti+HDh8PX1xcpKSnS9bi4OPTo0QNNmjRB9+7dsX79+qfeq6qqCn5+ftBoNLC0tES3bt3wxRdfSNcjIyOxZcsW/Oc//5Fm1D/99FOtywipqano378/lEolWrdujbCwMDx8+FC67uXlhaCgIMybNw8tWrSASqVCZGSkVj2RkZFo164dlEolnJycEBQU9OIDRi8N/l8zNVgXLlzAvn37YG5uDgCIjY1FREQE1q5di759+yIjIwP+/v6wtrbG+++/X+P91dXVaNu2Lb799lu0bNkSaWlp+PDDD9G6dWuMHz8eH330EfLy8lBWVoa4uDgAj85Pu379utZ9rl27hjfeeANTpkzB1q1b8fvvv8Pf3x9NmjTRCtQtW7YgJCQEx48fx7FjxzBlyhR4eHhg2LBh+O6777Bq1SokJCSgV69eKCoqwm+//Wa4wSPjM8qzxYj04P333xeNGzcW1tbWokmTJtIjG2NiYoQQQqjVarFjxw6t9yxevFgMHDhQCCFEQUGBACAyMjKe+hkBAQHi7bff1vrMt956S6vPk/f5+OOPRbdu3UR1dbXUZ926daJp06aiqqpKCPHoEYlPnvzar18/MX/+fCGEECtXrhRdu3YVlZWVug8INShcRqAGZejQocjMzMTx48cRGBiIESNGIDAwEDdu3MCVK1fg5+eHpk2bSq8lS5bg/PnzT73fV199BXd3d7Rq1QpNmzZFbGysdIyKrvLy8jBw4ECtc6w8PDxw9+5dXL16VWpzcXHRel/r1q1RXFwMAHjnnXdQXl6Ojh07wt/fH4mJiVrLENTwMWypQbG2tkbnzp3h4uKCNWvWoKKiAgsXLkR1dTWAR0sJ/3ucTE5ODn799dda7/Xtt99i7ty5+OCDD5CSkoLMzExMnToVlZWVsmoStRwYKP7/Y6L/t/3xcsdjCoVCqlutViM/Px/r1q2DpaUlAgICMGTIEDx48EBWLfTy4potNWgREREYNWoUZs6ciTZt2uDChQuYOHGiTu/95ZdfMGjQIK3dDE/Ogi0sLFBVVfXM+/Ts2RM7d+7UCt20tDTY2NigTZs2Ov8ulpaWGDNmDMaMGYNZs2ahe/fuyM7Ohqurq873oJcXw5YaNC8vL/Tq1QtRUVGIjIxEUFAQbG1tMWrUKFRUVODUqVMoKSlBSEhIjfd27twZW7duRXJyMjQaDbZt24aTJ09qHUzZoUMHJCcnIz8/H/b29rCzs6txn4CAAKxevRqBgYGYPXs28vPzERERgZCQEDRqpNt/PMbHx6OqqgoDBgyAlZUVtm3bBktLS+mwQmr4uIxADV5ISAhiY2MxYsQIbNq0CfHx8ejduzc8PT0RHx9f41Tfx2bMmIFx48bB19cXAwYMwK1bt2rs2fX390e3bt2kdd2jR4/WuE+bNm2wZ88enDhxAq+99hpmzJgBPz8//N///Z/Ov0OzZs0QGxsLDw8PuLi44ODBg0hKSoK9vb28waCXFs8gIyIyAs5siYiMgGFLRGQEDFsiIiNg2BIRGQHDlojICBi2RERGwLAlIjIChi0RkREwbImIjIBhS0RkBAxbIiIj+H8tVeDBB6pflQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 350x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data['relation_id'] = pd.factorize(data['relation'])[0]\n",
    "colors = ['orange']\n",
    "plt.figure(figsize=(7/2, 5/2))\n",
    "# Plot the histogram\n",
    "plt.hist(data['relation_id'], bins=len(data['relation'].unique()), align='left', color=colors)\n",
    "\n",
    "#plt.title('Distribution of Relations')\n",
    "plt.xlabel('Relations')\n",
    "plt.ylabel('Number of Sentences')\n",
    "\n",
    "# Save the figure as a PDF\n",
    "plt.savefig('images2/fewrel_relation_distribution.pdf', bbox_inches='tight',dpi=300)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "id": "ee88986e-5bbb-4e5e-86e1-09e9d2c2ae32",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import ast\n",
    "def numEntities(data):\n",
    "    data['entities'] = data['entities'].apply(ast.literal_eval)\n",
    "    data['tuple_length'] = data['entities'].apply(len)\n",
    "\n",
    "# Calculate the total number of values across all tuples\n",
    "    data_sum = data['tuple_length'].sum()\n",
    "    return data_sum/len(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5ab97646-d160-4568-9182-c8cb6b3fdf6d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 196,
   "id": "44f6feff-fd0f-42af-b3ee-5de33de94ef7",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import nltk\n",
    "from nltk.tokenize import word_tokenize\n",
    "\n",
    "# Sample DataFrame with a 'sentences' column\n",
    "def claculateTokens(df):\n",
    "\n",
    "    # Tokenize the sentences and calculate the number of tokens for each sentence\n",
    "    df['num_tokens'] = df['sentence'].apply(lambda x: len(word_tokenize(x)))\n",
    "\n",
    "    # Calculate the average number of tokens across all sentences\n",
    "    average_tokens = df['num_tokens'].mean()\n",
    "    print(\"\\nAverage number of tokens across all sentences:\", average_tokens)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "id": "37cb0bdf-237b-4f7a-b68a-bf581eb6fa03",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 219,
   "id": "aa45df6d-ca7c-4a67-8539-540e69a9e446",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Sample DataFrame with a 'sentences' column\n",
    "\n",
    "def avgCosineSim(df):\n",
    "    # Create a CountVectorizer to convert sentences to vectors\n",
    "    vectorizer = CountVectorizer()\n",
    "    X = vectorizer.fit_transform(df['sentence'])\n",
    "\n",
    "    # Calculate cosine similarity matrix\n",
    "    cosine_sim_matrix = cosine_similarity(X, X)\n",
    "\n",
    "    # Convert the similarity matrix to DataFrame for better visualization\n",
    "    cosine_sim_df = pd.DataFrame(cosine_sim_matrix, columns=df.index, index=df.index)\n",
    "\n",
    "    # Calculate the average cosine similarity excluding self-similarity\n",
    "    average_cosine_similarity = cosine_sim_df.mask(cosine_sim_df == 1.0).mean().mean()\n",
    "    print(\"\\nAverage cosine similarity between sentences:\", average_cosine_similarity)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e396570a-14bb-4a27-9ef3-6e7715925cdc",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}