categorize.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Entity type clustering\n",
    "Note: Only for demonstration purposes - categorizations used in training are created while running main.py, using code from categorize.py. The code here has been revised since, however the categorizations created are representative of the final one, except for OOD clustering. Actual categories vary by random state."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from copy import deepcopy\n",
    "from sklearn.cluster import AffinityPropagation\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.pipeline import Pipeline\n",
    "from gensim import downloader\n",
    "from tabulate import tabulate\n",
    "from dotenv import load_dotenv\n",
    "\n",
    "import nltk\n",
    "nltk.download('wordnet')\n",
    "nltk.download('omw-1.4') # Required for wordnet either in newer versions or on windows, i'm not sure which\n",
    "from nltk.corpus import wordnet\n",
    "\n",
    "from umap import UMAP\n",
    "from hdbscan import HDBSCAN\n",
    "\n",
    "\n",
    "import categorize\n",
    "\n",
    "load_dotenv();"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Manual Clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "manual_categories = {\n",
    "    \"person\" :      [\"politician\", \"person\", \"writer\", \"researcher\", \"scientist\", \"musicalartist\"],\n",
    "    \"organisation\": [\"organisation\", \"politicalparty\", \"university\", \"band\"],\n",
    "    \"location\":     [\"country\", \"location\"],\n",
    "    \"event\":        [\"event\", \"election\", \"award\", \"conference\"],\n",
    "    \"artifact\":     [\"album\", \"song\", \"academicjournal\", \"poem\", \"magazine\", \"book\"],\n",
    "    \"scientific\":   [\"metrics\", \"enzyme\", \"protein\", \"chemicalcompound\", \"chemicalelement\", \"astronomicalobject\"],\n",
    "    \"concept\":      [\"theory\", \"musicgenre\", \"field\", \"discipline\", \"algorithm\", \"literarygenre\"],\n",
    "    \"brand\":        [\"product\", \"programlang\"],\n",
    "    \"misc\":         [\"misc\", \"musicalinstrument\", \"task\"]\n",
    "}\n",
    "\n",
    "elisa_categories = {\n",
    "    \"person\":       [\"politician\", \"person\", \"writer\", \"researcher\", \"scientist\", \"musicalartist\"],\n",
    "    \"organisation\": [\"organisation\", \"politicalparty\", \"university\", \"band\"],\n",
    "    \"location\":     [\"country\", \"location\"],\n",
    "    \"event\":        [\"event\", \"election\", \"conference\"],\n",
    "    \"misc\":         [\"award\", \"album\", \"song\", \"academicjournal\", \"poem\", \"magazine\", \"book\", \"metrics\", \"enzyme\", \"protein\", \"chemicalcompound\", \"chemicalelement\", \"astronomicalobject\", \"theory\", \"musicgenre\", \"field\", \"discipline\", \"algorithm\", \"literarygenre\", \"product\", \"programlang\", \"misc\", \"musicalinstrument\", \"task\"]\n",
    "}"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Embedding based - Word2Vec"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Include all entity labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading pretrained Word2Vec model, this may take a while.\n",
      "These entities are not in the model:\n",
      "['academicjournal', 'astronomicalobject', 'chemicalcompound', 'chemicalelement', 'literarygenre', 'musicalartist', 'musicalinstrument', 'musicgenre', 'organisation', 'politicalparty', 'programlang']\n"
     ]
    }
   ],
   "source": [
    "# Avoid downloading / loading the whole Word2Vec Google News 300 model\n",
    "# If True, category names won't be calculated and files won't be written.\n",
    "quick = False\n",
    "random_state = 4012\n",
    "category_map_csv_path = f\"data/category_mappings/category_mapping_rs{random_state}.csv\"\n",
    "embeddings_csv_path = \"data/embeddings.csv\"\n",
    "\n",
    "# Labels (Manual copy from .env file)\n",
    "entities_orig = os.getenv(f\"ENTITY_LABELS\").split()\n",
    "\n",
    "if quick:\n",
    "    entities = ['journal', 'album', 'algorithm', 'galaxy', 'award', 'band', 'book', 'chemical', 'conference', 'country', 'discipline', 'election', 'enzyme', 'event', 'field', 'genre', 'location', 'magazine', 'metrics', 'miscellaneous', 'musician', 'violin', 'organization', 'person', 'poem', 'coalition', 'politician', 'product', 'javascript', 'protein', 'researcher', 'scientist', 'song', 'task', 'theory', 'university', 'writer']\n",
    "    embeddings = np.loadtxt(embeddings_csv_path, delimiter=\",\", dtype=\"float32\")\n",
    "    print(\"3D plot and category list will be created, but w/o category names and files won't be written.\")\n",
    "else:\n",
    "\n",
    "    # Load Word2Vec embeddings - First time download: ~1.6 GB\n",
    "    print(\"Loading pretrained Word2Vec model, this may take a while.\")\n",
    "    w2v = downloader.load(\"word2vec-google-news-300\")\n",
    "\n",
    "    # Check which words are not in Word2Vec model\n",
    "    missing = categorize.find_missing(w2v, entities_orig)\n",
    "    print(f\"These entities are not in the model:\\n{missing}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All entities are in the model. Final list:\n",
      "\n",
      "['journal', 'album', 'algorithm', 'galaxy', 'award', 'band', 'book', 'chemical', 'conference', 'country', 'discipline', 'election', 'enzyme', 'event', 'field', 'genre', 'location', 'magazine', 'metrics', 'miscellaneous', 'musician', 'violin', 'organization', 'person', 'poem', 'coalition', 'politician', 'product', 'javascript', 'protein', 'researcher', 'scientist', 'song', 'task', 'theory', 'university', 'writer']\n",
      "\n",
      "Loading embeddings for them.\n",
      "Done. Saving embeddings.\n"
     ]
    }
   ],
   "source": [
    "if not quick:\n",
    "    # Manual correction\n",
    "    substitute = {\"musicalartist\": \"musician\", \"organisation\": \"organization\", \"politicalparty\": \"coalition\", \"academicjournal\": \"journal\", \"chemicalcompound\": \"chemical\", \"chemicalelement\": \"chemical\", \"astronomicalobject\": \"galaxy\", \"musicgenre\": \"genre\", \"literarygenre\": \"genre\", \"programlang\": \"javascript\", \"musicalinstrument\": \"violin\", \"misc\": \"miscellaneous\"}\n",
    "    entities = []\n",
    "    for entity in entities_orig:\n",
    "        if entity in substitute:\n",
    "            to_add = substitute[entity]\n",
    "        else:\n",
    "            to_add = entity\n",
    "        if to_add not in entities:\n",
    "            entities.append(to_add)\n",
    "\n",
    "    still_missing = categorize.find_missing(w2v, entities)\n",
    "    if len(still_missing) > 0:\n",
    "        print(\"These entities are not in the model:\")\n",
    "        print(still_missing)\n",
    "    else:\n",
    "        print(\"All entities are in the model. Final list:\\n\")\n",
    "        print(entities)\n",
    "        print(\"\\nLoading embeddings for them.\")\n",
    "        embeddings = w2v[entities]\n",
    "        print(\"Done. Saving embeddings.\")\n",
    "        np.savetxt(embeddings_csv_path, embeddings, delimiter=\",\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2eeef5447e9945b2befa1065fcc0ba78",
       "version_major": 2,
       "version_minor": 0
      },
      "image/png": "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",
      "text/html": [
       "\n",
       "            <div style=\"display: inline-block;\">\n",
       "                <div class=\"jupyter-widgets widget-label\" style=\"text-align: center;\">\n",
       "                    Figure\n",
       "                </div>\n",
       "                <img src='data:image/png;base64,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' width=900.0/>\n",
       "            </div>\n",
       "        "
      ],
      "text/plain": [
       "Canvas(toolbar=Toolbar(toolitems=[('Home', 'Reset original view', 'home', 'home'), ('Back', 'Back to previous …"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Calculating label names, this might take a few sec.\n",
      "\n",
      "The final categories are:\n",
      "----------  -------------------------------------------------------------------------------------------------------------------------\n",
      "essay       ['academicjournal', 'book', 'magazine', 'poem', 'theory']\n",
      "songs       ['album', 'band', 'musicgenre', 'literarygenre', 'musicalinstrument', 'song']\n",
      "proteins    ['algorithm', 'astronomicalobject', 'chemicalcompound', 'chemicalelement', 'enzyme', 'metrics', 'programlang', 'protein']\n",
      "convention  ['award', 'conference', 'election', 'event', 'location', 'organisation', 'person', 'politicalparty']\n",
      "TELW        ['country', 'discipline', 'field', 'misc', 'product', 'task', 'university']\n",
      "journalist  ['musicalartist', 'politician', 'researcher', 'scientist', 'writer']\n",
      "----------  -------------------------------------------------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "# Tuned vars\n",
    "n_neighbors = 4\n",
    "min_dist = 0.3\n",
    "damping = 0.5\n",
    "\n",
    "pipe = Pipeline([\n",
    "    (\"PCA\", PCA(n_components=35, svd_solver=\"full\")),\n",
    "    (\"UMAP\", UMAP(n_neighbors=n_neighbors, n_components=3, metric=\"cosine\", min_dist=min_dist, random_state=random_state)),\n",
    "    (\"Clusterer\", AffinityPropagation(damping=damping, random_state=random_state))\n",
    "])\n",
    "pred = pipe.fit_predict(embeddings)\n",
    "\n",
    "# Plot\n",
    "color_palette = sns.color_palette()\n",
    "cluster_colors = [color_palette[x] for x in pred]\n",
    "%matplotlib widget\n",
    "fig = plt.figure(figsize=(9, 6))\n",
    "ax = plt.axes(projection='3d')\n",
    "coordinates = pipe[0:2].transform(embeddings)\n",
    "plot = ax.scatter3D(coordinates[:,0], coordinates[:,1], coordinates[:,2], c=cluster_colors, s=50, linewidth=0)\n",
    "ax.set_title(\"Word2Vec clustering\")\n",
    "ax.set_xlabel('x')\n",
    "ax.set_ylabel('y')\n",
    "ax.set_zlabel('z')\n",
    "plt.show()\n",
    "\n",
    "# Category names\n",
    "categories = categorize.categories_from_pred(pred, entities)\n",
    "# for fun\n",
    "categories_oldnames = deepcopy(categories)\n",
    "if not quick:\n",
    "    print(\"\\nCalculating label names, this might take a few sec.\")\n",
    "    categories = categorize.entities_to_orig(categorize.name_categories(categories, w2v), substitute)\n",
    "    embedding_categories = categories\n",
    "else:\n",
    "    print(\"Set `quick` to False to get named categories. Defaulting to unnamed labels.\")\n",
    "    embedding_categories = None\n",
    "print(\"\\nThe final categories are:\")\n",
    "print(tabulate([[category, elements] for category, elements in categories.items()]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is mostly for fun because I liked the quirky names of the old way I did the category naming.\n",
      "\n",
      "-----------------------  -------------------------------------------------------------------------------------------------------------------------\n",
      "essays                   ['academicjournal', 'book', 'magazine', 'poem', 'theory']\n",
      "tunes                    ['album', 'band', 'musicgenre', 'literarygenre', 'musicalinstrument', 'song']\n",
      "molecules                ['algorithm', 'astronomicalobject', 'chemicalcompound', 'chemicalelement', 'enzyme', 'metrics', 'programlang', 'protein']\n",
      "executive_Ellen_Ziffren  ['award', 'conference', 'election', 'event', 'location', 'organisation', 'person', 'politicalparty']\n",
      "XKSD_What                ['country', 'discipline', 'field', 'misc', 'product', 'task', 'university']\n",
      "historian                ['musicalartist', 'politician', 'researcher', 'scientist', 'writer']\n",
      "-----------------------  -------------------------------------------------------------------------------------------------------------------------\n",
      "-----------------------------------------------  -------------------------------------------------------------------------------------------------------------------------\n",
      "journals                                         ['academicjournal', 'book', 'magazine', 'poem', 'theory']\n",
      "music                                            ['album', 'band', 'musicgenre', 'literarygenre', 'musicalinstrument', 'song']\n",
      "metabolic_enzymes                                ['algorithm', 'astronomicalobject', 'chemicalcompound', 'chemicalelement', 'enzyme', 'metrics', 'programlang', 'protein']\n",
      "NetWitness_NextGen_contact_sales@netwitness.com  ['award', 'conference', 'election', 'event', 'location', 'organisation', 'person', 'politicalparty']\n",
      "graduate_assistantships                          ['country', 'discipline', 'field', 'misc', 'product', 'task', 'university']\n",
      "filmmaker                                        ['musicalartist', 'politician', 'researcher', 'scientist', 'writer']\n",
      "-----------------------------------------------  -------------------------------------------------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "# For fun: with the old naming function (avoids too similar names + doesn't remove names with special chars)\n",
    "if not quick:\n",
    "    print(\"This is mostly for fun because I liked the quirky names of the old way I did the category naming.\\n\")\n",
    "    categories_5th = categorize.entities_to_orig(categorize.name_categories_old(categories_oldnames, w2v, kth_closest=5), substitute)\n",
    "    categories_6th = categorize.entities_to_orig(categorize.name_categories_old(categories_oldnames, w2v, kth_closest=6), substitute)\n",
    "    print(tabulate([[category, elements] for category, elements in categories_5th.items()]))\n",
    "    print(tabulate([[category, elements] for category, elements in categories_6th.items()]))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### OOD clustering - OLD!!\n",
    "Note: Does not include unique label exclusion!!\n",
    "\n",
    "Cluster based on 5 domain, predict categories of 6th domain\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting processing of test domain AI.\n",
      "  The test entities (that are unique to this domain) are:\n",
      "  ['algorithm', 'conference', 'field', 'metrics', 'product', 'researcher', 'task']\n",
      "AI done.\n",
      "\n",
      "Starting processing of test domain LITERATURE.\n",
      "  The test entities (that are unique to this domain) are:\n",
      "  ['poem', 'writer']\n",
      "LITERATURE done.\n",
      "\n",
      "Starting processing of test domain MUSIC.\n",
      "  The test entities (that are unique to this domain) are:\n",
      "  []\n",
      "MUSIC done.\n",
      "\n",
      "Starting processing of test domain NEWS.\n",
      "  The test entities (that are unique to this domain) are:\n",
      "  []\n",
      "NEWS done.\n",
      "\n",
      "Starting processing of test domain POLITICS.\n",
      "  The test entities (that are unique to this domain) are:\n",
      "  ['coalition', 'election', 'politician']\n",
      "POLITICS done.\n",
      "\n",
      "Starting processing of test domain SCIENCE.\n",
      "  The test entities (that are unique to this domain) are:\n",
      "  ['chemical', 'discipline', 'enzyme', 'galaxy', 'protein', 'theory']\n",
      "SCIENCE done.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "if not quick:\n",
    "    # Get dict of entities for each domain + substitute\n",
    "    domains = [\"ai\", \"literature\", \"music\", \"news\", \"politics\", \"science\"]\n",
    "    entity_dict = {}\n",
    "    for domain in domains:\n",
    "        entities = set()\n",
    "        for entity in os.getenv(f\"{domain.upper()}_LABELS\").split():\n",
    "            if entity in substitute:\n",
    "                entities.add(substitute[entity])\n",
    "            else:\n",
    "                entities.add(entity)\n",
    "        entity_dict[domain] = sorted(list(entities))\n",
    "    \n",
    "    # Do a clustering for each OOD possibility\n",
    "    ood_categories = {}\n",
    "    for test_domain in domains:\n",
    "        print(f\"Starting processing of test domain {test_domain.upper()}.\")\n",
    "        # List of all entities not in the test domain\n",
    "        train_entities = set()\n",
    "        [train_entities.update(domain_entities) for domain, domain_entities in entity_dict.items() if domain != test_domain]\n",
    "        train_entities = sorted(list(train_entities))\n",
    "        # And for those only in the test domain\n",
    "        test_entities = [entity for entity in entity_dict[test_domain] if entity not in train_entities]\n",
    "        print(f\"  The test entities (that are unique to this domain) are:\\n  {test_entities}\")\n",
    "        # Get embeddings for them\n",
    "        train_embeddings = w2v[train_entities]\n",
    "        # Only if there are entitites unique to the test domain\n",
    "        if len(test_entities) > 0:\n",
    "            test_embeddings = w2v[test_entities]\n",
    "\n",
    "        # Fit the pipeline\n",
    "        # Max nr. of components for PCA: min(n_samples, n_features). We selected 35\n",
    "        # for the global clustering (with 37 entities in the substituted list).\n",
    "        # Here we'll have less entitites -> Limits the nr of components we can have.\n",
    "        n_components = min(len(train_entities), 35)\n",
    "        pipe = Pipeline([\n",
    "            (\"PCA\", PCA(n_components=n_components, svd_solver=\"full\")),\n",
    "            (\"UMAP\", UMAP(n_neighbors=n_neighbors, n_components=3, metric=\"cosine\", min_dist=min_dist, random_state=random_state)),\n",
    "            (\"Clusterer\", AffinityPropagation(damping=damping, random_state=random_state))\n",
    "        ])\n",
    "        train_pred = pipe.fit_predict(train_embeddings)\n",
    "        # Predict cluster of test domain entitites (if there are any)\n",
    "        if len(test_entities) > 0:\n",
    "            test_pred = pipe.predict(test_embeddings)\n",
    "            # Concat everything and continue like before\n",
    "            pred = list(train_pred) + list(test_pred)\n",
    "            entities = train_entities + test_entities\n",
    "        else:\n",
    "            pred = train_pred\n",
    "            entities = train_entities\n",
    "        categories = categorize.categories_from_pred(pred, entities)\n",
    "        ood_categories[test_domain] = categorize.entities_to_orig(categorize.name_categories(categories, w2v), substitute)\n",
    "        print(f\"{test_domain.upper()} done.\\n\")\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Thesaurus-based categories - WordNet"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Topological"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['album', 'algorithm', 'award', 'band', 'book', 'chemical', 'conference', 'country', 'discipline', 'election', 'enzyme', 'event', 'field', 'galaxy', 'genre', 'instrument', 'java', 'journal', 'location', 'magazine', 'metrics', 'miscellaneous', 'musician', 'organization', 'party', 'person', 'poem', 'politician', 'product', 'protein', 'researcher', 'scientist', 'song', 'task', 'theory', 'university', 'writer']\n"
     ]
    }
   ],
   "source": [
    "# Manual Correction\n",
    "substitute = {\"musicalartist\": \"musician\", \"organisation\": \"organization\", \"politicalparty\": \"party\", \"academicjournal\": \"journal\", \"chemicalcompound\": \"chemical\", \"chemicalelement\": \"chemical\", \"astronomicalobject\": \"galaxy\", \"musicgenre\": \"genre\", \"literarygenre\": \"genre\", \"programlang\": \"java\", \"musicalinstrument\": \"instrument\", \"misc\": \"miscellaneous\"}\n",
    "entities = set()\n",
    "for entity in entities_orig:\n",
    "    if entity in substitute:\n",
    "        entities.add(substitute[entity])\n",
    "    else:\n",
    "        entities.add(entity)\n",
    "\n",
    "entities = sorted(list(entities))\n",
    "print(entities)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get categories with most frequent word meanings\n",
    "# categorize.category_dict(most_frequent_synset(entities), 2, entities)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "36 writer\n",
      "0 Synset('writer.n.01') writes (books or stories or articles or the like) professionally (for pay)\n",
      "1 Synset('writer.n.02') a person who is able to write and has written something\n"
     ]
    }
   ],
   "source": [
    "# Check meanings and definitions for manual correction\n",
    "idx = 36\n",
    "synlist = wordnet.synsets(entities[idx])\n",
    "print(idx, entities[idx])\n",
    "for i, syn in enumerate(synlist):\n",
    "    print(i, syn, syn.definition())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Manual corrections of meanings. Key: idx in Entity_labels, value: idx of meaning in wordnet.synsets(word)\n",
    "# check: To check whether entities are the same as the list the meaning dict was made on\n",
    "check = ['album', 'algorithm', 'award', 'band', 'book', 'chemical', 'conference', 'country', 'discipline', 'election', 'enzyme', 'event', 'field', 'galaxy', 'genre', 'instrument', 'java', 'journal', 'location', 'magazine', 'metrics', 'miscellaneous', 'musician', 'organization', 'party', 'person', 'poem', 'politician', 'product', 'protein', 'researcher', 'scientist', 'song', 'task', 'theory', 'university', 'writer']\n",
    "if entities != check:\n",
    "    raise ValueError(\"entities modified since meaning_dict was edited!\")\n",
    "meaning_dict = {0: 0, 1: 0, 2: 1, 3: 0, 4: 0, 5: 0, 6: 0, 7: 0, 8: 0, 9: 0, 10: 0, 11: 0, 12: 3, 13: 2, 14: 0, 15: 5, 16: 2, 17: 1, 18: 0, 19: 0, 20: 3, 21: 0, 22: 0, 23: 2, 24: 0, 25: 0, 26: 0, 27: 0, 28: 0, 29: 0, 30: 0, 31: 0, 32: 0, 33: 1, 34: 0, 35: 2, 36: 0}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "The final categories are:\n",
      "---------------------  ----------------------------------------------------------------------------------------------------------\n",
      "object                 ['album', 'book', 'musicalinstrument', 'academicjournal', 'location', 'magazine', 'product']\n",
      "psychological_feature  ['algorithm', 'discipline', 'election', 'event', 'field', 'musicgenre', 'literarygenre', 'task', 'theory']\n",
      "communication          ['award', 'programlang', 'poem', 'song']\n",
      "group                  ['band', 'conference', 'country', 'astronomicalobject', 'organisation', 'politicalparty', 'university']\n",
      "matter                 ['chemicalcompound', 'chemicalelement', 'enzyme']\n",
      "measure                ['metrics']\n",
      "assorted               ['misc']\n",
      "causal_agent           ['musicalartist', 'person', 'politician', 'researcher', 'scientist', 'writer']\n",
      "thing                  ['protein']\n",
      "---------------------  ----------------------------------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "# Get categories with associated words\n",
    "entity_syns = []\n",
    "for i, word in enumerate(entities):\n",
    "    entity_syns.append(wordnet.synsets(word)[meaning_dict[i]])\n",
    "topological_categories = categorize.topological_category_dict(entity_syns, 2, entities)\n",
    "topological_categories = categorize.entities_to_orig(topological_categories, substitute)\n",
    "print(\"\\nThe final categories are:\")\n",
    "print(tabulate([[category, elements] for category, elements in topological_categories.items()]))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Affinity-based"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Calculating label names, this might take a few sec.\n",
      "\n",
      "The final categories are:\n",
      "----------  -----------------------------------------------------------------------------------------------------------------------------\n",
      "journals    ['album', 'book', 'musicalinstrument', 'academicjournal', 'magazine', 'product']\n",
      "pleiotropy  ['algorithm', 'discipline', 'election', 'event', 'field', 'musicgenre', 'literarygenre', 'metrics', 'misc', 'task', 'theory']\n",
      "poems       ['award', 'programlang', 'poem', 'song']\n",
      "group       ['band', 'conference', 'country', 'astronomicalobject', 'organisation', 'politicalparty', 'university']\n",
      "enzymes     ['chemicalcompound', 'chemicalelement', 'enzyme', 'protein']\n",
      "journalist  ['location', 'musicalartist', 'person', 'politician', 'researcher', 'scientist', 'writer']\n",
      "----------  -----------------------------------------------------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "# Calculate Wu-Palmer Similarities\n",
    "similarities = categorize.get_wup_sim(entity_syns)\n",
    "# Affinity Propagation\n",
    "damping = 0.5 # [0.5, 1.0)\n",
    "ap = AffinityPropagation(damping=damping, affinity=\"precomputed\", random_state=random_state)\n",
    "ap = ap.fit_predict(similarities)\n",
    "\n",
    "# Category names\n",
    "categories = categorize.categories_from_pred(ap, entities)\n",
    "# for fun\n",
    "categories_oldnames = deepcopy(categories)\n",
    "if not quick:\n",
    "    print(\"\\nCalculating label names, this might take a few sec.\")\n",
    "    categories = categorize.entities_to_orig(categorize.name_categories(categories, w2v), substitute)\n",
    "    thesaurus_similarity_categories = categories\n",
    "else:\n",
    "    print(\"Set `quick` to False to get named categories. Defaulting to unnamed labels.\")\n",
    "    thesaurus_similarity_categories = None\n",
    "print(\"\\nThe final categories are:\")\n",
    "print(tabulate([[category, elements] for category, elements in categories.items()]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is mostly for fun because I liked the quirky names of the old way I did the category naming.\n",
      "\n",
      "------------------  -----------------------------------------------------------------------------------------------------------------------------\n",
      "anthology           ['album', 'book', 'musicalinstrument', 'academicjournal', 'magazine', 'product']\n",
      "By_Tatyana_Shumsky  ['algorithm', 'discipline', 'election', 'event', 'field', 'musicgenre', 'literarygenre', 'metrics', 'misc', 'task', 'theory']\n",
      "lyrics              ['award', 'programlang', 'poem', 'song']\n",
      "Hal_Kratzsch        ['band', 'conference', 'country', 'astronomicalobject', 'organisation', 'politicalparty', 'university']\n",
      "biochemical         ['chemicalcompound', 'chemicalelement', 'enzyme', 'protein']\n",
      "songwriter          ['location', 'musicalartist', 'person', 'politician', 'researcher', 'scientist', 'writer']\n",
      "------------------  -----------------------------------------------------------------------------------------------------------------------------\n",
      "-----------------  -----------------------------------------------------------------------------------------------------------------------------\n",
      "CD                 ['album', 'book', 'musicalinstrument', 'academicjournal', 'magazine', 'product']\n",
      "resonated_Calkins  ['algorithm', 'discipline', 'election', 'event', 'field', 'musicgenre', 'literarygenre', 'metrics', 'misc', 'task', 'theory']\n",
      "Poem               ['award', 'programlang', 'poem', 'song']\n",
      "nation             ['band', 'conference', 'country', 'astronomicalobject', 'organisation', 'politicalparty', 'university']\n",
      "cysteine           ['chemicalcompound', 'chemicalelement', 'enzyme', 'protein']\n",
      "poet               ['location', 'musicalartist', 'person', 'politician', 'researcher', 'scientist', 'writer']\n",
      "-----------------  -----------------------------------------------------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "# For fun: with the old naming function (avoids too similar names + doesn't remove names with special chars)\n",
    "if not quick:\n",
    "    print(\"This is mostly for fun because I liked the quirky names of the old way I did the category naming.\\n\")\n",
    "    categories_5th = categorize.entities_to_orig(categorize.name_categories_old(categories_oldnames, w2v, kth_closest=5), substitute)\n",
    "    categories_6th = categorize.entities_to_orig(categorize.name_categories_old(categories_oldnames, w2v, kth_closest=6), substitute)\n",
    "    print(tabulate([[category, elements] for category, elements in categories_5th.items()]))\n",
    "    print(tabulate([[category, elements] for category, elements in categories_6th.items()]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Entities categorized as noise:\n",
      "['location', 'metrics', 'miscellaneous']\n",
      "\n",
      "Calculating label names, this might take a few sec.\n",
      "\n",
      "The final categories are:\n",
      "-----------  ----------------------------------------------------------------------------------------------------------\n",
      "tome         ['album', 'book', 'academicjournal', 'magazine']\n",
      "pedagogies   ['algorithm', 'discipline', 'election', 'event', 'field', 'musicgenre', 'literarygenre', 'task', 'theory']\n",
      "poems        ['award', 'programlang', 'poem', 'song']\n",
      "group        ['band', 'conference', 'country', 'astronomicalobject', 'organisation', 'politicalparty', 'university']\n",
      "enzymes      ['chemicalcompound', 'chemicalelement', 'enzyme', 'protein']\n",
      "instruments  ['musicalinstrument', 'product']\n",
      "locations    ['location', 'metrics', 'misc']\n",
      "journalist   ['musicalartist', 'person', 'politician', 'researcher', 'scientist', 'writer']\n",
      "-----------  ----------------------------------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "# This was just a test - We can use it if we can argue why this over Affinity Propagation\n",
    "# Propagation if we used that for the word2vec\n",
    "distances = 1 - similarities\n",
    "hdb = HDBSCAN(min_cluster_size=2, min_samples=2, cluster_selection_epsilon=0, metric=\"precomputed\")\n",
    "hdb = hdb.fit_predict(distances)\n",
    "\n",
    "categories = categorize.categories_from_pred(hdb, entities)\n",
    "# for fun\n",
    "categories_oldnames = deepcopy(categories)\n",
    "if -1 in categories:\n",
    "    print(f\"Entities categorized as noise:\\n{categories[-1]}\")\n",
    "if not quick:\n",
    "    print(\"\\nCalculating label names, this might take a few sec.\")\n",
    "    categories = categorize.entities_to_orig(categorize.name_categories(categories, w2v), substitute)\n",
    "    hdb_thesaurus_similarity_categories = categories\n",
    "else:\n",
    "    print(\"Set `quick` to False to get named categories. Defaulting to unnamed labels.\")\n",
    "    hdb_thesaurus_similarity_categories = None\n",
    "print(\"\\nThe final categories are:\")\n",
    "print(tabulate([[category, elements] for category, elements in categories.items()]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is mostly for fun because I liked the quirky names of the old way I did the category naming.\n",
      "\n",
      "---------------  ----------------------------------------------------------------------------------------------------------\n",
      "autobiography    ['album', 'book', 'academicjournal', 'magazine']\n",
      "unfair_Fredline  ['algorithm', 'discipline', 'election', 'event', 'field', 'musicgenre', 'literarygenre', 'task', 'theory']\n",
      "lyrics           ['award', 'programlang', 'poem', 'song']\n",
      "Hal_Kratzsch     ['band', 'conference', 'country', 'astronomicalobject', 'organisation', 'politicalparty', 'university']\n",
      "biochemical      ['chemicalcompound', 'chemicalelement', 'enzyme', 'protein']\n",
      "SourceMeter      ['musicalinstrument', 'product']\n",
      "TrimFleet        ['location', 'metrics', 'misc']\n",
      "physicist        ['musicalartist', 'person', 'politician', 'researcher', 'scientist', 'writer']\n",
      "---------------  ----------------------------------------------------------------------------------------------------------\n",
      "--------------------------  ----------------------------------------------------------------------------------------------------------\n",
      "Magazine                    ['album', 'book', 'academicjournal', 'magazine']\n",
      "By_Tatyana_Shumsky          ['algorithm', 'discipline', 'election', 'event', 'field', 'musicgenre', 'literarygenre', 'task', 'theory']\n",
      "Poem                        ['award', 'programlang', 'poem', 'song']\n",
      "nation                      ['band', 'conference', 'country', 'astronomicalobject', 'organisation', 'politicalparty', 'university']\n",
      "cysteine                    ['chemicalcompound', 'chemicalelement', 'enzyme', 'protein']\n",
      "electromagnetic_flowmeter   ['musicalinstrument', 'product']\n",
      "performance_indicators_KPI  ['location', 'metrics', 'misc']\n",
      "poet                        ['musicalartist', 'person', 'politician', 'researcher', 'scientist', 'writer']\n",
      "--------------------------  ----------------------------------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "# For fun: with the old naming function (avoids too similar names + doesn't remove names with special chars)\n",
    "if not quick:\n",
    "    print(\"This is mostly for fun because I liked the quirky names of the old way I did the category naming.\\n\")\n",
    "    categories_5th = categorize.entities_to_orig(categorize.name_categories_old(categories_oldnames, w2v, kth_closest=5), substitute)\n",
    "    categories_6th = categorize.entities_to_orig(categorize.name_categories_old(categories_oldnames, w2v, kth_closest=6), substitute)\n",
    "    print(tabulate([[category, elements] for category, elements in categories_5th.items()]))\n",
    "    print(tabulate([[category, elements] for category, elements in categories_6th.items()]))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Write to file\n",
    "Not used anymore - main.py creates categorizations upon calling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # Only if category names are assigned\n",
    "# if not quick:\n",
    "#     # check if everything matches\n",
    "#     entities_orig.sort()\n",
    "#     category_types = [manual_categories, elisa_categories, embedding_categories, topological_categories, thesaurus_similarity_categories] + [ood_categories[test_domain] for test_domain in domains]\n",
    "#     for category_type in category_types:\n",
    "#         tmp = []\n",
    "#         for values in category_type.values():\n",
    "#             tmp += values\n",
    "#         tmp.sort()\n",
    "#         if tmp != entities_orig:\n",
    "#             raise ValueError(\"Entities don't match across categorization methods.\")\n",
    "\n",
    "#     # Dict with entities as keys and the list of category names as values\n",
    "#     # corresponding to the order in category_types\n",
    "#     entity_dict = {}\n",
    "#     for category_type in category_types:\n",
    "#         for category, entities in category_type.items():\n",
    "#             for entity in entities:\n",
    "#                 if entity in entity_dict:\n",
    "#                     entity_dict[entity].append(category)\n",
    "#                 else:\n",
    "#                     entity_dict[entity] = [category]\n",
    "    \n",
    "#     cols = [\"manual\", \"elisa\", \"embedding\", \"topological\", \"thesaurus_affinity\"] + [f\"ood_embedding_{domain}\" for domain in domains]\n",
    "#     category_df = pd.DataFrame.from_dict(entity_dict, orient=\"index\", columns=cols)\n",
    "#     category_df.index.name = \"entity_name\"\n",
    "#     os.makedirs(os.path.dirname(category_map_csv_path), exist_ok=True)\n",
    "#     category_df.to_csv(category_map_csv_path)\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "2yp",
   "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": "3.9.16"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}