{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "cc2aed6b-5a72-4950-8f64-aad5e252eb90",
   "metadata": {},
   "source": [
    "# Identify Entity Names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "583e9d12-d82e-4a9e-aa95-3b3d1c45d549",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "38456 42330\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd, json, numpy as np\n",
    "\n",
    "wcfile = \"../data/wiki/simplewiki-20211120/wordcount.min10.json\"\n",
    "countfile = \"../data/wiki/simplewiki-20211120/count.min10.json\"\n",
    "evalfile = \"evaluation/Mewsli-9/en.tsv\"\n",
    "\n",
    "# wcfile = 'wiki/nlwiki-20220301/wordcount.min2.json'\n",
    "# countfile = 'wiki/nlwiki-20220301/count.min2.json'\n",
    "# evalfile = 'evaluation/Mewsli-9/nl.tsv'\n",
    "\n",
    "\n",
    "# Load word counts\n",
    "wc = pd.read_json(wcfile, orient=\"index\")[0].sort_values()\n",
    "# Load anchor-entity counts\n",
    "aec = json.load(open(countfile))\n",
    "aec = pd.DataFrame([(a, e, c) for a, ec in aec.items() for e, c in ec.items()])\n",
    "aec[1] = aec[1].str.replace(\"Q\", \"\").astype(int)\n",
    "\n",
    "# Load bad entities\n",
    "badent = pd.read_csv(\"data/wikidata-20211122-disambig+list.txt\", header=None)[0]\n",
    "# aec = aec[~aec[1].isin(badent)]\n",
    "aec = aec.set_index([0, 1])[2]\n",
    "print(len(wc), len(aec))\n",
    "\n",
    "\n",
    "import pandas as pd, json, numpy as np\n",
    "from minimel import normalize\n",
    "\n",
    "ev = pd.read_csv(evalfile, header=None, sep=\"\\t\")[1]\n",
    "ev = ev.map(lambda x: [n for surf in json.loads(x) for n in normalize(surf)]).explode()\n",
    "\n",
    "df = pd.DataFrame(\n",
    "    {\"word_count\": wc, \"link_count\": aec.groupby(level=0).sum()}, dtype=\"Int64\"\n",
    ").dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d32699f1-375d-48dc-a856-c6380da6308c",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df[\"ratio\"] = np.log(df.word_count) / np.log(df.link_count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "5f35fa77-414d-4872-b574-7044bd40ce16",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Counting eval words: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 250/250 [00:00<00:00, 1307.94it/s]\n"
     ]
    }
   ],
   "source": [
    "import tqdm\n",
    "from minimel.mentions import count_name_lines\n",
    "\n",
    "lines = dict(zip(list(range(250)), open(evalfile))).values()\n",
    "eval_count = dict(\n",
    "    count_name_lines(tqdm.tqdm(lines, desc=\"Counting eval words\"), countfile)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8a7d0290-6411-4de0-bccf-7335f67e79a4",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='ratio', ylabel='eval_count'>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='ratio'>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "We want a trade-off between accepting bad entities and rejecting good ones.\n",
    "\"\"\"\n",
    "df2 = df.copy()\n",
    "df2[\"eval_count\"] = pd.Series(eval_count).astype(\"Int64\")\n",
    "df2[\"eval_linked\"] = df2.index.isin(ev)\n",
    "df2.dropna(inplace=True)\n",
    "df2.sort_values(\"ratio\", inplace=True)\n",
    "df2\n",
    "display(\n",
    "    df2.plot.scatter(\n",
    "        x=\"ratio\",\n",
    "        y=\"eval_count\",\n",
    "        c=df2[\"eval_linked\"].map(lambda x: \"g\" if x else \"r\"),\n",
    "        logy=True,\n",
    "    )\n",
    ")\n",
    "\n",
    "s = pd.DataFrame(\n",
    "    {\n",
    "        \"false positives\": (df2[\"eval_count\"] * ~df2[\"eval_linked\"]).cumsum(),\n",
    "        \"false negatives\": (df2[\"eval_count\"] * df2[\"eval_linked\"])[::-1].cumsum()[\n",
    "            ::-1\n",
    "        ],\n",
    "        \"true positives\": (df2[\"eval_count\"] * df2[\"eval_linked\"]).cumsum(),\n",
    "        \"ratio\": df2[\"ratio\"],\n",
    "    }\n",
    ")\n",
    "s[\"precision\"] = s[\"true positives\"] / (s[\"true positives\"] + s[\"false positives\"])\n",
    "s[\"recall\"] = s[\"true positives\"] / (s[\"true positives\"] + s[\"false negatives\"])\n",
    "s[\"f1\"] = 2 * (s[\"precision\"] * s[\"recall\"]) / (s[\"precision\"] + s[\"recall\"])\n",
    "s.set_index(\"ratio\")[[\"precision\", \"recall\", \"f1\"]].plot.line()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "42521df9-ac9a-4572-8d68-af6ca63f9ee7",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Threshold: 1.951297634655119\n",
      "keeping 36813 ; dropping 1642\n",
      "dropping eval names: 261\n",
      "{'2 december', 'patch', 'iraqi', 'target', 'maria', '26 april', 'first', 'judgment', 'dogs', 'tate', 'paralympics', 'reagan', 'kilometers', 'east coast', 'politically', 'reboot', 'unity', 'decatur', 'fusion', 'practice', 'association', 'teachers', 'english-language', 'south american', 'shadow', 'berlin, germany', 'de', 'wheelchair basketball', 'it', 'laurel', 'landmarks', 'the age', 'friendly', 'james', 'reduced', '25 june', 'out', 'civilians', 'jack', 'crustaceans', 'parents', 'prime', 'nelson', 'freestyle', 'in', 'awake', 'elementary', 'ace', 'buses', 'clinton', 'ron', 'red river', 'clouds', 'beta', 'communications', 'inches', 'kilometres', 'scarecrow', 'registered', 'guardian', 'values', 'dudley', 'pigs', 'licensing', 'shock', 's', 'cheyenne', 'offensive', 'bo', 'respect', 'minor', 'apples', 'the moon', 'golfer', 'irving', 'daniel', 'john', 'kenyan', 'uprising', 'lynn', 'heinz', 'left', 'scientists', 'metres', '25 september', 'run', 'flash', 'draft', 'meeting', 'bolivian', 'the game', 'saint john', 'agent', 'swift', 'astronomers', 'patrick', 'present', 'resolution', 'syrian', 'hope', 'focus', 'trophy', 'tornadoes', 'neutrality', 'png', 'europeans', 'concord', 'ministers', 'father', 'sharks', 'sri lankan', 'drivers', 'best actress', 'public service', 'northwest', 'north korean', 'wells', 'kilograms', 'this', 'commission', 'lost', 'core', 'lp', 'the left', 'acres', 'jackson', 'angels', 'apes', 'corona', 'saints', 'augusta', 'masters', 'convention', 'medium', 'w', 'unions', 'value', 'evening', 'eight', 'duty', 'refugees', 'mark', 'who', 'notre dame', 'believe', 'wayne', 'british house of commons', 'converted', 'lifetime', 'staff', 'thomas', 'leadership', 'manor', 'triple crown', 'may', 'nico', 'farmers', 'clyde', 'taxes', 'fairy tales', 'ted', 'bishops', 'freud', '30 november', 'create', 'lighting', 'svg', 'novels', '21 march', 'tortured', 'today', 'colombian', 'x', 'moe', 'led', 'malaysian', 'partnership', 'jefferson', 'tie', 'trained', 'seven', 'capitals', 'southern', 'rabbits', 'short', 'museums', 'arrested', 'physicists', 'ministry', 'editorial', 'runner', 'mg', 'presidential election', 'bullets', 'criticized', 'retailer', 'sale', 'protests', 'psychologists', 'general assembly', 'b.c.', 'trailer', 'presidency', 'straight', 'edison', 'crash', 'divisions', 'the edge', 'roosevelt', 'stanford', 'al', 'tna', 'oceans', 'dolphins', 'hearts', 'show', 'algerian', 'path', 'constitutional', 'vista', 'council', 'loop', 'manuscripts', 'miles', 'opposition', 'riding', 'civil', 'runs', 'anc', 'wife', 'blacks', 'spiders', 'helicopters', 'bangladeshi', 'elections', 'evolutionary', 'run out', 'cornell', 'the white house', 'paul', 'engagement', 'false', 'style', 'elizabeth', 'complex', 'id', 'outbreak', 'speak', 'destiny', 'again', 'peter', 'north american', 'firefighters', 'frederick', 'lying', 'if', 'study', 'moderate', 'holly', 'websites', 'dollars'}\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div style=\"display:flex; flex-direction: row;\">\n",
       "        <style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_d1872\">\n",
       "  <caption>False Positives</caption>\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_d1872_level0_col0\" class=\"col_heading level0 col0\" >word_count</th>\n",
       "      <th id=\"T_d1872_level0_col1\" class=\"col_heading level0 col1\" >link_count</th>\n",
       "      <th id=\"T_d1872_level0_col2\" class=\"col_heading level0 col2\" >ratio</th>\n",
       "      <th id=\"T_d1872_level0_col3\" class=\"col_heading level0 col3\" >eval_count</th>\n",
       "      <th id=\"T_d1872_level0_col4\" class=\"col_heading level0 col4\" >eval_linked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row0\" class=\"row_heading level0 row0\" >year</th>\n",
       "      <td id=\"T_d1872_row0_col0\" class=\"data row0 col0\" >22166</td>\n",
       "      <td id=\"T_d1872_row0_col1\" class=\"data row0 col1\" >454</td>\n",
       "      <td id=\"T_d1872_row0_col2\" class=\"data row0 col2\" >1.635527</td>\n",
       "      <td id=\"T_d1872_row0_col3\" class=\"data row0 col3\" >140</td>\n",
       "      <td id=\"T_d1872_row0_col4\" class=\"data row0 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row1\" class=\"row_heading level0 row1\" >can</th>\n",
       "      <td id=\"T_d1872_row1_col0\" class=\"data row1 col0\" >50542</td>\n",
       "      <td id=\"T_d1872_row1_col1\" class=\"data row1 col1\" >337</td>\n",
       "      <td id=\"T_d1872_row1_col2\" class=\"data row1 col2\" >1.860894</td>\n",
       "      <td id=\"T_d1872_row1_col3\" class=\"data row1 col3\" >111</td>\n",
       "      <td id=\"T_d1872_row1_col4\" class=\"data row1 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row2\" class=\"row_heading level0 row2\" >state</th>\n",
       "      <td id=\"T_d1872_row2_col0\" class=\"data row2 col0\" >19243</td>\n",
       "      <td id=\"T_d1872_row2_col1\" class=\"data row2 col1\" >4312</td>\n",
       "      <td id=\"T_d1872_row2_col2\" class=\"data row2 col2\" >1.178721</td>\n",
       "      <td id=\"T_d1872_row2_col3\" class=\"data row2 col3\" >80</td>\n",
       "      <td id=\"T_d1872_row2_col4\" class=\"data row2 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row3\" class=\"row_heading level0 row3\" >day</th>\n",
       "      <td id=\"T_d1872_row3_col0\" class=\"data row3 col0\" >14248</td>\n",
       "      <td id=\"T_d1872_row3_col1\" class=\"data row3 col1\" >325</td>\n",
       "      <td id=\"T_d1872_row3_col2\" class=\"data row3 col2\" >1.653641</td>\n",
       "      <td id=\"T_d1872_row3_col3\" class=\"data row3 col3\" >64</td>\n",
       "      <td id=\"T_d1872_row3_col4\" class=\"data row3 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row4\" class=\"row_heading level0 row4\" >party</th>\n",
       "      <td id=\"T_d1872_row4_col0\" class=\"data row4 col0\" >7020</td>\n",
       "      <td id=\"T_d1872_row4_col1\" class=\"data row4 col1\" >142</td>\n",
       "      <td id=\"T_d1872_row4_col2\" class=\"data row4 col2\" >1.787092</td>\n",
       "      <td id=\"T_d1872_row4_col3\" class=\"data row4 col3\" >64</td>\n",
       "      <td id=\"T_d1872_row4_col4\" class=\"data row4 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row5\" class=\"row_heading level0 row5\" >u.s.</th>\n",
       "      <td id=\"T_d1872_row5_col0\" class=\"data row5 col0\" >4676</td>\n",
       "      <td id=\"T_d1872_row5_col1\" class=\"data row5 col1\" >2064</td>\n",
       "      <td id=\"T_d1872_row5_col2\" class=\"data row5 col2\" >1.107148</td>\n",
       "      <td id=\"T_d1872_row5_col3\" class=\"data row5 col3\" >63</td>\n",
       "      <td id=\"T_d1872_row5_col4\" class=\"data row5 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row6\" class=\"row_heading level0 row6\" >world</th>\n",
       "      <td id=\"T_d1872_row6_col0\" class=\"data row6 col0\" >19029</td>\n",
       "      <td id=\"T_d1872_row6_col1\" class=\"data row6 col1\" >524</td>\n",
       "      <td id=\"T_d1872_row6_col2\" class=\"data row6 col2\" >1.573702</td>\n",
       "      <td id=\"T_d1872_row6_col3\" class=\"data row6 col3\" >63</td>\n",
       "      <td id=\"T_d1872_row6_col4\" class=\"data row6 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row7\" class=\"row_heading level0 row7\" >group</th>\n",
       "      <td id=\"T_d1872_row7_col0\" class=\"data row7 col0\" >17128</td>\n",
       "      <td id=\"T_d1872_row7_col1\" class=\"data row7 col1\" >269</td>\n",
       "      <td id=\"T_d1872_row7_col2\" class=\"data row7 col2\" >1.742444</td>\n",
       "      <td id=\"T_d1872_row7_col3\" class=\"data row7 col3\" >59</td>\n",
       "      <td id=\"T_d1872_row7_col4\" class=\"data row7 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row8\" class=\"row_heading level0 row8\" >second</th>\n",
       "      <td id=\"T_d1872_row8_col0\" class=\"data row8 col0\" >15927</td>\n",
       "      <td id=\"T_d1872_row8_col1\" class=\"data row8 col1\" >190</td>\n",
       "      <td id=\"T_d1872_row8_col2\" class=\"data row8 col2\" >1.844049</td>\n",
       "      <td id=\"T_d1872_row8_col3\" class=\"data row8 col3\" >51</td>\n",
       "      <td id=\"T_d1872_row8_col4\" class=\"data row8 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row9\" class=\"row_heading level0 row9\" >country</th>\n",
       "      <td id=\"T_d1872_row9_col0\" class=\"data row9 col0\" >12390</td>\n",
       "      <td id=\"T_d1872_row9_col1\" class=\"data row9 col1\" >4411</td>\n",
       "      <td id=\"T_d1872_row9_col2\" class=\"data row9 col2\" >1.123070</td>\n",
       "      <td id=\"T_d1872_row9_col3\" class=\"data row9 col3\" >49</td>\n",
       "      <td id=\"T_d1872_row9_col4\" class=\"data row9 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row10\" class=\"row_heading level0 row10\" >million</th>\n",
       "      <td id=\"T_d1872_row10_col0\" class=\"data row10 col0\" >9756</td>\n",
       "      <td id=\"T_d1872_row10_col1\" class=\"data row10 col1\" >615</td>\n",
       "      <td id=\"T_d1872_row10_col2\" class=\"data row10 col2\" >1.430423</td>\n",
       "      <td id=\"T_d1872_row10_col3\" class=\"data row10 col3\" >46</td>\n",
       "      <td id=\"T_d1872_row10_col4\" class=\"data row10 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row11\" class=\"row_heading level0 row11\" >fire</th>\n",
       "      <td id=\"T_d1872_row11_col0\" class=\"data row11 col0\" >4523</td>\n",
       "      <td id=\"T_d1872_row11_col1\" class=\"data row11 col1\" >570</td>\n",
       "      <td id=\"T_d1872_row11_col2\" class=\"data row11 col2\" >1.326412</td>\n",
       "      <td id=\"T_d1872_row11_col3\" class=\"data row11 col3\" >44</td>\n",
       "      <td id=\"T_d1872_row11_col4\" class=\"data row11 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row12\" class=\"row_heading level0 row12\" >public</th>\n",
       "      <td id=\"T_d1872_row12_col0\" class=\"data row12 col0\" >5622</td>\n",
       "      <td id=\"T_d1872_row12_col1\" class=\"data row12 col1\" >220</td>\n",
       "      <td id=\"T_d1872_row12_col2\" class=\"data row12 col2\" >1.600860</td>\n",
       "      <td id=\"T_d1872_row12_col3\" class=\"data row12 col3\" >43</td>\n",
       "      <td id=\"T_d1872_row12_col4\" class=\"data row12 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row13\" class=\"row_heading level0 row13\" >interview</th>\n",
       "      <td id=\"T_d1872_row13_col0\" class=\"data row13 col0\" >957</td>\n",
       "      <td id=\"T_d1872_row13_col1\" class=\"data row13 col1\" >116</td>\n",
       "      <td id=\"T_d1872_row13_col2\" class=\"data row13 col2\" >1.443920</td>\n",
       "      <td id=\"T_d1872_row13_col3\" class=\"data row13 col3\" >43</td>\n",
       "      <td id=\"T_d1872_row13_col4\" class=\"data row13 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row14\" class=\"row_heading level0 row14\" >man</th>\n",
       "      <td id=\"T_d1872_row14_col0\" class=\"data row14 col0\" >9036</td>\n",
       "      <td id=\"T_d1872_row14_col1\" class=\"data row14 col1\" >169</td>\n",
       "      <td id=\"T_d1872_row14_col2\" class=\"data row14 col2\" >1.775663</td>\n",
       "      <td id=\"T_d1872_row14_col3\" class=\"data row14 col3\" >42</td>\n",
       "      <td id=\"T_d1872_row14_col4\" class=\"data row14 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row15\" class=\"row_heading level0 row15\" >uk</th>\n",
       "      <td id=\"T_d1872_row15_col0\" class=\"data row15 col0\" >5383</td>\n",
       "      <td id=\"T_d1872_row15_col1\" class=\"data row15 col1\" >947</td>\n",
       "      <td id=\"T_d1872_row15_col2\" class=\"data row15 col2\" >1.253557</td>\n",
       "      <td id=\"T_d1872_row15_col3\" class=\"data row15 col3\" >42</td>\n",
       "      <td id=\"T_d1872_row15_col4\" class=\"data row15 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row16\" class=\"row_heading level0 row16\" >week</th>\n",
       "      <td id=\"T_d1872_row16_col0\" class=\"data row16 col0\" >2683</td>\n",
       "      <td id=\"T_d1872_row16_col1\" class=\"data row16 col1\" >75</td>\n",
       "      <td id=\"T_d1872_row16_col2\" class=\"data row16 col2\" >1.828538</td>\n",
       "      <td id=\"T_d1872_row16_col3\" class=\"data row16 col3\" >39</td>\n",
       "      <td id=\"T_d1872_row16_col4\" class=\"data row16 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row17\" class=\"row_heading level0 row17\" >question</th>\n",
       "      <td id=\"T_d1872_row17_col0\" class=\"data row17 col0\" >1430</td>\n",
       "      <td id=\"T_d1872_row17_col1\" class=\"data row17 col1\" >62</td>\n",
       "      <td id=\"T_d1872_row17_col2\" class=\"data row17 col2\" >1.760405</td>\n",
       "      <td id=\"T_d1872_row17_col3\" class=\"data row17 col3\" >36</td>\n",
       "      <td id=\"T_d1872_row17_col4\" class=\"data row17 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row18\" class=\"row_heading level0 row18\" >set</th>\n",
       "      <td id=\"T_d1872_row18_col0\" class=\"data row18 col0\" >10241</td>\n",
       "      <td id=\"T_d1872_row18_col1\" class=\"data row18 col1\" >176</td>\n",
       "      <td id=\"T_d1872_row18_col2\" class=\"data row18 col2\" >1.785936</td>\n",
       "      <td id=\"T_d1872_row18_col3\" class=\"data row18 col3\" >35</td>\n",
       "      <td id=\"T_d1872_row18_col4\" class=\"data row18 col4\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d1872_level0_row19\" class=\"row_heading level0 row19\" >information</th>\n",
       "      <td id=\"T_d1872_row19_col0\" class=\"data row19 col0\" >5795</td>\n",
       "      <td id=\"T_d1872_row19_col1\" class=\"data row19 col1\" >329</td>\n",
       "      <td id=\"T_d1872_row19_col2\" class=\"data row19 col2\" >1.494939</td>\n",
       "      <td id=\"T_d1872_row19_col3\" class=\"data row19 col3\" >34</td>\n",
       "      <td id=\"T_d1872_row19_col4\" class=\"data row19 col4\" >False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "&nbsp;&nbsp;<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_d8ec8\">\n",
       "  <caption>False Negatives</caption>\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_d8ec8_level0_col0\" class=\"col_heading level0 col0\" >word_count</th>\n",
       "      <th id=\"T_d8ec8_level0_col1\" class=\"col_heading level0 col1\" >link_count</th>\n",
       "      <th id=\"T_d8ec8_level0_col2\" class=\"col_heading level0 col2\" >ratio</th>\n",
       "      <th id=\"T_d8ec8_level0_col3\" class=\"col_heading level0 col3\" >eval_count</th>\n",
       "      <th id=\"T_d8ec8_level0_col4\" class=\"col_heading level0 col4\" >eval_linked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row0\" class=\"row_heading level0 row0\" >in</th>\n",
       "      <td id=\"T_d8ec8_row0_col0\" class=\"data row0 col0\" >882984</td>\n",
       "      <td id=\"T_d8ec8_row0_col1\" class=\"data row0 col1\" >193</td>\n",
       "      <td id=\"T_d8ec8_row0_col2\" class=\"data row0 col2\" >2.601533</td>\n",
       "      <td id=\"T_d8ec8_row0_col3\" class=\"data row0 col3\" >1855</td>\n",
       "      <td id=\"T_d8ec8_row0_col4\" class=\"data row0 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row1\" class=\"row_heading level0 row1\" >s</th>\n",
       "      <td id=\"T_d8ec8_row1_col0\" class=\"data row1 col0\" >144874</td>\n",
       "      <td id=\"T_d8ec8_row1_col1\" class=\"data row1 col1\" >364</td>\n",
       "      <td id=\"T_d8ec8_row1_col2\" class=\"data row1 col2\" >2.015145</td>\n",
       "      <td id=\"T_d8ec8_row1_col3\" class=\"data row1 col3\" >692</td>\n",
       "      <td id=\"T_d8ec8_row1_col4\" class=\"data row1 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row2\" class=\"row_heading level0 row2\" >it</th>\n",
       "      <td id=\"T_d8ec8_row2_col0\" class=\"data row2 col0\" >237945</td>\n",
       "      <td id=\"T_d8ec8_row2_col1\" class=\"data row2 col1\" >43</td>\n",
       "      <td id=\"T_d8ec8_row2_col2\" class=\"data row2 col2\" >3.291448</td>\n",
       "      <td id=\"T_d8ec8_row2_col3\" class=\"data row2 col3\" >495</td>\n",
       "      <td id=\"T_d8ec8_row2_col4\" class=\"data row2 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row3\" class=\"row_heading level0 row3\" >this</th>\n",
       "      <td id=\"T_d8ec8_row3_col0\" class=\"data row3 col0\" >95628</td>\n",
       "      <td id=\"T_d8ec8_row3_col1\" class=\"data row3 col1\" >158</td>\n",
       "      <td id=\"T_d8ec8_row3_col2\" class=\"data row3 col2\" >2.265285</td>\n",
       "      <td id=\"T_d8ec8_row3_col3\" class=\"data row3 col3\" >383</td>\n",
       "      <td id=\"T_d8ec8_row3_col4\" class=\"data row3 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row4\" class=\"row_heading level0 row4\" >who</th>\n",
       "      <td id=\"T_d8ec8_row4_col0\" class=\"data row4 col0\" >44973</td>\n",
       "      <td id=\"T_d8ec8_row4_col1\" class=\"data row4 col1\" >38</td>\n",
       "      <td id=\"T_d8ec8_row4_col2\" class=\"data row4 col2\" >2.945310</td>\n",
       "      <td id=\"T_d8ec8_row4_col3\" class=\"data row4 col3\" >223</td>\n",
       "      <td id=\"T_d8ec8_row4_col4\" class=\"data row4 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row5\" class=\"row_heading level0 row5\" >first</th>\n",
       "      <td id=\"T_d8ec8_row5_col0\" class=\"data row5 col0\" >60542</td>\n",
       "      <td id=\"T_d8ec8_row5_col1\" class=\"data row5 col1\" >13</td>\n",
       "      <td id=\"T_d8ec8_row5_col2\" class=\"data row5 col2\" >4.292908</td>\n",
       "      <td id=\"T_d8ec8_row5_col3\" class=\"data row5 col3\" >140</td>\n",
       "      <td id=\"T_d8ec8_row5_col4\" class=\"data row5 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row6\" class=\"row_heading level0 row6\" >if</th>\n",
       "      <td id=\"T_d8ec8_row6_col0\" class=\"data row6 col0\" >27312</td>\n",
       "      <td id=\"T_d8ec8_row6_col1\" class=\"data row6 col1\" >25</td>\n",
       "      <td id=\"T_d8ec8_row6_col2\" class=\"data row6 col2\" >3.173493</td>\n",
       "      <td id=\"T_d8ec8_row6_col3\" class=\"data row6 col3\" >119</td>\n",
       "      <td id=\"T_d8ec8_row6_col4\" class=\"data row6 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row7\" class=\"row_heading level0 row7\" >out</th>\n",
       "      <td id=\"T_d8ec8_row7_col0\" class=\"data row7 col0\" >19446</td>\n",
       "      <td id=\"T_d8ec8_row7_col1\" class=\"data row7 col1\" >63</td>\n",
       "      <td id=\"T_d8ec8_row7_col2\" class=\"data row7 col2\" >2.383557</td>\n",
       "      <td id=\"T_d8ec8_row7_col3\" class=\"data row7 col3\" >116</td>\n",
       "      <td id=\"T_d8ec8_row7_col4\" class=\"data row7 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row8\" class=\"row_heading level0 row8\" >today</th>\n",
       "      <td id=\"T_d8ec8_row8_col0\" class=\"data row8 col0\" >6340</td>\n",
       "      <td id=\"T_d8ec8_row8_col1\" class=\"data row8 col1\" >36</td>\n",
       "      <td id=\"T_d8ec8_row8_col2\" class=\"data row8 col2\" >2.443027</td>\n",
       "      <td id=\"T_d8ec8_row8_col3\" class=\"data row8 col3\" >74</td>\n",
       "      <td id=\"T_d8ec8_row8_col4\" class=\"data row8 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row9\" class=\"row_heading level0 row9\" >may</th>\n",
       "      <td id=\"T_d8ec8_row9_col0\" class=\"data row9 col0\" >20704</td>\n",
       "      <td id=\"T_d8ec8_row9_col1\" class=\"data row9 col1\" >132</td>\n",
       "      <td id=\"T_d8ec8_row9_col2\" class=\"data row9 col2\" >2.035324</td>\n",
       "      <td id=\"T_d8ec8_row9_col3\" class=\"data row9 col3\" >51</td>\n",
       "      <td id=\"T_d8ec8_row9_col4\" class=\"data row9 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row10\" class=\"row_heading level0 row10\" >al</th>\n",
       "      <td id=\"T_d8ec8_row10_col0\" class=\"data row10 col0\" >4833</td>\n",
       "      <td id=\"T_d8ec8_row10_col1\" class=\"data row10 col1\" >26</td>\n",
       "      <td id=\"T_d8ec8_row10_col2\" class=\"data row10 col2\" >2.603736</td>\n",
       "      <td id=\"T_d8ec8_row10_col3\" class=\"data row10 col3\" >44</td>\n",
       "      <td id=\"T_d8ec8_row10_col4\" class=\"data row10 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row11\" class=\"row_heading level0 row11\" >left</th>\n",
       "      <td id=\"T_d8ec8_row11_col0\" class=\"data row11 col0\" >11344</td>\n",
       "      <td id=\"T_d8ec8_row11_col1\" class=\"data row11 col1\" >75</td>\n",
       "      <td id=\"T_d8ec8_row11_col2\" class=\"data row11 col2\" >2.162471</td>\n",
       "      <td id=\"T_d8ec8_row11_col3\" class=\"data row11 col3\" >31</td>\n",
       "      <td id=\"T_d8ec8_row11_col4\" class=\"data row11 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row12\" class=\"row_heading level0 row12\" >again</th>\n",
       "      <td id=\"T_d8ec8_row12_col0\" class=\"data row12 col0\" >7952</td>\n",
       "      <td id=\"T_d8ec8_row12_col1\" class=\"data row12 col1\" >17</td>\n",
       "      <td id=\"T_d8ec8_row12_col2\" class=\"data row12 col2\" >3.169962</td>\n",
       "      <td id=\"T_d8ec8_row12_col3\" class=\"data row12 col3\" >29</td>\n",
       "      <td id=\"T_d8ec8_row12_col4\" class=\"data row12 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row13\" class=\"row_heading level0 row13\" >show</th>\n",
       "      <td id=\"T_d8ec8_row13_col0\" class=\"data row13 col0\" >11522</td>\n",
       "      <td id=\"T_d8ec8_row13_col1\" class=\"data row13 col1\" >70</td>\n",
       "      <td id=\"T_d8ec8_row13_col2\" class=\"data row13 col2\" >2.201253</td>\n",
       "      <td id=\"T_d8ec8_row13_col3\" class=\"data row13 col3\" >26</td>\n",
       "      <td id=\"T_d8ec8_row13_col4\" class=\"data row13 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row14\" class=\"row_heading level0 row14\" >run</th>\n",
       "      <td id=\"T_d8ec8_row14_col0\" class=\"data row14 col0\" >5137</td>\n",
       "      <td id=\"T_d8ec8_row14_col1\" class=\"data row14 col1\" >13</td>\n",
       "      <td id=\"T_d8ec8_row14_col2\" class=\"data row14 col2\" >3.331147</td>\n",
       "      <td id=\"T_d8ec8_row14_col3\" class=\"data row14 col3\" >26</td>\n",
       "      <td id=\"T_d8ec8_row14_col4\" class=\"data row14 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row15\" class=\"row_heading level0 row15\" >believe</th>\n",
       "      <td id=\"T_d8ec8_row15_col0\" class=\"data row15 col0\" >3406</td>\n",
       "      <td id=\"T_d8ec8_row15_col1\" class=\"data row15 col1\" >14</td>\n",
       "      <td id=\"T_d8ec8_row15_col2\" class=\"data row15 col2\" >3.081894</td>\n",
       "      <td id=\"T_d8ec8_row15_col3\" class=\"data row15 col3\" >24</td>\n",
       "      <td id=\"T_d8ec8_row15_col4\" class=\"data row15 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row16\" class=\"row_heading level0 row16\" >elections</th>\n",
       "      <td id=\"T_d8ec8_row16_col0\" class=\"data row16 col0\" >1471</td>\n",
       "      <td id=\"T_d8ec8_row16_col1\" class=\"data row16 col1\" >31</td>\n",
       "      <td id=\"T_d8ec8_row16_col2\" class=\"data row16 col2\" >2.123973</td>\n",
       "      <td id=\"T_d8ec8_row16_col3\" class=\"data row16 col3\" >23</td>\n",
       "      <td id=\"T_d8ec8_row16_col4\" class=\"data row16 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row17\" class=\"row_heading level0 row17\" >arrested</th>\n",
       "      <td id=\"T_d8ec8_row17_col0\" class=\"data row17 col0\" >1569</td>\n",
       "      <td id=\"T_d8ec8_row17_col1\" class=\"data row17 col1\" >13</td>\n",
       "      <td id=\"T_d8ec8_row17_col2\" class=\"data row17 col2\" >2.868748</td>\n",
       "      <td id=\"T_d8ec8_row17_col3\" class=\"data row17 col3\" >23</td>\n",
       "      <td id=\"T_d8ec8_row17_col4\" class=\"data row17 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row18\" class=\"row_heading level0 row18\" >meeting</th>\n",
       "      <td id=\"T_d8ec8_row18_col0\" class=\"data row18 col0\" >1389</td>\n",
       "      <td id=\"T_d8ec8_row18_col1\" class=\"data row18 col1\" >27</td>\n",
       "      <td id=\"T_d8ec8_row18_col2\" class=\"data row18 col2\" >2.195600</td>\n",
       "      <td id=\"T_d8ec8_row18_col3\" class=\"data row18 col3\" >22</td>\n",
       "      <td id=\"T_d8ec8_row18_col4\" class=\"data row18 col4\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d8ec8_level0_row19\" class=\"row_heading level0 row19\" >john</th>\n",
       "      <td id=\"T_d8ec8_row19_col0\" class=\"data row19 col0\" >12391</td>\n",
       "      <td id=\"T_d8ec8_row19_col1\" class=\"data row19 col1\" >63</td>\n",
       "      <td id=\"T_d8ec8_row19_col2\" class=\"data row19 col2\" >2.274781</td>\n",
       "      <td id=\"T_d8ec8_row19_col3\" class=\"data row19 col3\" >20</td>\n",
       "      <td id=\"T_d8ec8_row19_col4\" class=\"data row19 col4\" >True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "\n",
       "    </div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import HTML\n",
    "\n",
    "\n",
    "def display_row(captioned_dataframes):\n",
    "    return HTML(\n",
    "        f\"\"\"<div style=\"display:flex; flex-direction: row;\">\n",
    "        {\"&nbsp;&nbsp;\".join(df.style.set_caption(c)._repr_html_() for c,df in captioned_dataframes.items())}\n",
    "    </div>\"\"\"\n",
    "    )\n",
    "\n",
    "\n",
    "f1 = s.set_index(\"ratio\")[\"f1\"].astype('float').fillna(0)\n",
    "cutoff = f1[abs(f1 - f1.max()) < 0.05].index[0]\n",
    "print(\"Threshold:\", cutoff)\n",
    "print(f\"keeping\", len(df[df.ratio < cutoff]), \"; dropping\", len(df[df.ratio > cutoff]))\n",
    "\n",
    "drop_ev = set(ev) & set(df[df.ratio > cutoff].index)\n",
    "print(\"dropping eval names:\", len(drop_ev))\n",
    "print(drop_ev)\n",
    "\n",
    "display_row(\n",
    "    {\n",
    "        \"False Positives\": df2[~df2[\"eval_linked\"] & (df2[\"ratio\"] < cutoff)]\n",
    "        .sort_values(\"eval_count\")[::-1]\n",
    "        .head(20),\n",
    "        \"False Negatives\": df2[df2[\"eval_linked\"] & (df2[\"ratio\"] > cutoff)]\n",
    "        .sort_values(\"eval_count\")[::-1]\n",
    "        .head(20),\n",
    "    }\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f2fe0d48-ee39-4c18-a6a6-3e4dd1d67ba4",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from minimel.normalize import normalize\n",
    "from minimel.vectorize import vw_tok\n",
    "\n",
    "import json, math, tqdm, dawg\n",
    "\n",
    "surface_weights = json.load(open(countfile))\n",
    "\n",
    "# filter out selection\n",
    "drop_tok = set(tuple(vw_tok(d)) for d in df[df.ratio > cutoff].index)\n",
    "surface_weights = {\n",
    "    k: v\n",
    "    for k, v in surface_weights.items()\n",
    "    if (tuple(vw_tok(k)) not in drop_tok) and not k[-1] == \"-\"\n",
    "}\n",
    "\n",
    "surface_trie = dawg.CompletionDAWG(surface_weights)\n",
    "import pathlib\n",
    "\n",
    "surface_trie.save(\n",
    "    pathlib.Path(countfile).parent\n",
    "    / (pathlib.Path(countfile).stem + \".salient.completiondawg\")\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7a0b80e-ffee-4249-afd0-ea75864332bb",
   "metadata": {},
   "source": [
    "## Analyse Mention Detection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "0da31372-f0c7-429d-aab1-32b67315b313",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "'stille oceaan'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "File \u001b[0;32m~/.conda/envs/p38/lib/python3.8/site-packages/pandas/core/indexes/base.py:3652\u001b[0m, in \u001b[0;36mIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   3651\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 3652\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_engine\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_loc\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcasted_key\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   3653\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n",
      "File \u001b[0;32m~/.conda/envs/p38/lib/python3.8/site-packages/pandas/_libs/index.pyx:147\u001b[0m, in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
      "File \u001b[0;32m~/.conda/envs/p38/lib/python3.8/site-packages/pandas/_libs/index.pyx:176\u001b[0m, in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
      "File \u001b[0;32mpandas/_libs/hashtable_class_helper.pxi:7080\u001b[0m, in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
      "File \u001b[0;32mpandas/_libs/hashtable_class_helper.pxi:7088\u001b[0m, in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;31mKeyError\u001b[0m: 'stille oceaan'",
      "\nThe above exception was the direct cause of the following exception:\n",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[15], line 2\u001b[0m\n\u001b[1;32m      1\u001b[0m l \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mstille oceaan\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m----> 2\u001b[0m display(\u001b[43mdf\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mloc\u001b[49m\u001b[43m[\u001b[49m\u001b[43ml\u001b[49m\u001b[43m]\u001b[49m)\n\u001b[1;32m      3\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m l \u001b[38;5;129;01min\u001b[39;00m aec:\n\u001b[1;32m      4\u001b[0m     display(aec[l], l \u001b[38;5;129;01min\u001b[39;00m drop_ev)\n",
      "File \u001b[0;32m~/.conda/envs/p38/lib/python3.8/site-packages/pandas/core/indexing.py:1103\u001b[0m, in \u001b[0;36m_LocationIndexer.__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   1100\u001b[0m axis \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39maxis \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;241m0\u001b[39m\n\u001b[1;32m   1102\u001b[0m maybe_callable \u001b[38;5;241m=\u001b[39m com\u001b[38;5;241m.\u001b[39mapply_if_callable(key, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mobj)\n\u001b[0;32m-> 1103\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_getitem_axis\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmaybe_callable\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/.conda/envs/p38/lib/python3.8/site-packages/pandas/core/indexing.py:1343\u001b[0m, in \u001b[0;36m_LocIndexer._getitem_axis\u001b[0;34m(self, key, axis)\u001b[0m\n\u001b[1;32m   1341\u001b[0m \u001b[38;5;66;03m# fall thru to straight lookup\u001b[39;00m\n\u001b[1;32m   1342\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_validate_key(key, axis)\n\u001b[0;32m-> 1343\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_get_label\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/.conda/envs/p38/lib/python3.8/site-packages/pandas/core/indexing.py:1293\u001b[0m, in \u001b[0;36m_LocIndexer._get_label\u001b[0;34m(self, label, axis)\u001b[0m\n\u001b[1;32m   1291\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_get_label\u001b[39m(\u001b[38;5;28mself\u001b[39m, label, axis: AxisInt):\n\u001b[1;32m   1292\u001b[0m     \u001b[38;5;66;03m# GH#5567 this will fail if the label is not present in the axis.\u001b[39;00m\n\u001b[0;32m-> 1293\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mobj\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mxs\u001b[49m\u001b[43m(\u001b[49m\u001b[43mlabel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/.conda/envs/p38/lib/python3.8/site-packages/pandas/core/generic.py:4095\u001b[0m, in \u001b[0;36mNDFrame.xs\u001b[0;34m(self, key, axis, level, drop_level)\u001b[0m\n\u001b[1;32m   4093\u001b[0m             new_index \u001b[38;5;241m=\u001b[39m index[loc]\n\u001b[1;32m   4094\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 4095\u001b[0m     loc \u001b[38;5;241m=\u001b[39m \u001b[43mindex\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_loc\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   4097\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(loc, np\u001b[38;5;241m.\u001b[39mndarray):\n\u001b[1;32m   4098\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m loc\u001b[38;5;241m.\u001b[39mdtype \u001b[38;5;241m==\u001b[39m np\u001b[38;5;241m.\u001b[39mbool_:\n",
      "File \u001b[0;32m~/.conda/envs/p38/lib/python3.8/site-packages/pandas/core/indexes/base.py:3654\u001b[0m, in \u001b[0;36mIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   3652\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_engine\u001b[38;5;241m.\u001b[39mget_loc(casted_key)\n\u001b[1;32m   3653\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n\u001b[0;32m-> 3654\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01merr\u001b[39;00m\n\u001b[1;32m   3655\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m:\n\u001b[1;32m   3656\u001b[0m     \u001b[38;5;66;03m# If we have a listlike key, _check_indexing_error will raise\u001b[39;00m\n\u001b[1;32m   3657\u001b[0m     \u001b[38;5;66;03m#  InvalidIndexError. Otherwise we fall through and re-raise\u001b[39;00m\n\u001b[1;32m   3658\u001b[0m     \u001b[38;5;66;03m#  the TypeError.\u001b[39;00m\n\u001b[1;32m   3659\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_indexing_error(key)\n",
      "\u001b[0;31mKeyError\u001b[0m: 'stille oceaan'"
     ]
    }
   ],
   "source": [
    "l = \"stille oceaan\"\n",
    "display(df.loc[l])\n",
    "if l in aec:\n",
    "    display(aec[l], l in drop_ev)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "ead2a808-c5cf-4973-acc4-847d8060d438",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Q184425': 2, 'Q98': 1274}"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "surface_weights[\"stille oceaan\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "dad9bfa3-501e-40d7-bd51-2468a0edb337",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Zware aardbeving voor de kust in zuiden Mexico  8 september 2017   Zware aardbeving 8.2 voor de kust in zuiden Mexico  De zuidkust van Mexico is omstreeks middernacht plaatselijke tijd getroffen door een aardbeving. De Mexicaanse seismologische dienst zegt dat de beving een kracht van 8.2 op de schaal van Richter had. Het centrum was in de Stille Oceaan, op 70 kilometer diepte, zo'n 100 kilometer ten zuidwesten van Pijijiapan. De meeste doden vielen in Oaxaca en het aangrenzende Chiapas. Ook in buurland Guatemala heeft de beving een dodelijk slachtoffer geëist.  De aardbeving was ook te voelen in Mexico-Stad, waar mensen in paniek de straat op renden. De aardbeving zorgde even voor een grootschalige stroomuitval bij 1 miljoen mensen, die snel weer kon worden verholpen. De beving veroorzaakte ook een lichte tsunami.  Een dergelijke krachtige aardbeving komt gemiddeld één keer per jaar voor. De vorige van dit kaliber was in 2015, in Chili. De laatste zware aardbeving in Mexico (waarbij vele duizenden doden vielen) was in 1985. Deze nieuwe beving is echter volgens president Nieto de krachtigste in Mexico in 100 jaar.  Mexicaans president Enrique Peña Nieto heeft daags nadien drie dagen van nationale rouw afgekondigd.  Het officiële dodental staat nu op 65. Vooral in Juchitan is veel schade aan woningen.  Bronnen.  Zelf schrijven? Hoe schrijf ik een artikel?\n",
      "{'Oaxaca': 34110, 'Chiapas': 60123, 'Guatemala': 774, 'Mexico-Stad': 1489, 'in 1985': 1798567}\n",
      "zware aardbeving {'Q214866': 2, 'Q43777': 15, 'Q151835': 2, 'Q211386': 3, 'Q1798567': 2}\n",
      "aardbeving {'Q214866': 3, 'Q1903316': 3, 'Q749610': 3, 'Q1053476': 2, 'Q116688': 8, 'Q16068825': 2, 'Q191055': 5, 'Q328663': 7, 'Q210137': 2, 'Q2441400': 7, 'Q43777': 9, 'Q1909872': 2, 'Q19830062': 5, 'Q2885880': 2, 'Q3245204': 3, 'Q274498': 2, 'Q27850765': 2, 'Q2483366': 2, 'Q3161000': 2, 'Q38821': 3, 'Q939106': 6, 'Q1082836': 3, 'Q130754': 2, 'Q1847006': 2, 'Q1348954': 3, 'Q26689701': 3, 'Q745411': 3, 'Q151835': 4, 'Q1990627': 2, 'Q1348910': 5, 'Q211386': 7, 'Q7944': 1417, 'Q867691': 5, 'Q152033': 4, 'Q47144124': 2, 'Q207918': 5, 'Q1798567': 3}\n",
      "mexico {'Q588480': 2, 'Q260965': 6, 'Q15296784': 18, 'Q599923': 19, 'Q6797304': 4, 'Q285658': 4, 'Q170603': 3, 'Q55712': 11, 'Q641583': 3, 'Q50647': 3, 'Q96': 10551, 'Q2151952': 12, 'Q941143': 2, 'Q8109': 2, 'Q48311380': 2, 'Q108810124': 4, 'Q51636908': 3, 'Q1489': 105, 'Q478399': 6, 'Q82112': 263, 'Q371334': 16, 'Q2294975': 9, 'Q764690': 8, 'Q2776455': 6, 'Q29066592': 3, 'Q603181': 3, 'Q173099': 2, 'Q1182352': 11, 'Q1166178': 3, 'Q1926260': 10, 'Q17041307': 5, 'Q150644': 2, 'Q2342098': 7, 'Q2712776': 3, 'Q12123144': 3, 'Q22696198': 2, 'Q164089': 908, 'Q1160682': 22, 'Q1907023': 4, 'Q3098723': 4, 'Q279769': 2, 'Q1041675': 2, 'Q387588': 3, 'Q19826175': 13, 'Q178652': 12, 'Q2376913': 2, 'Q43196466': 9, 'Q1414354': 3, 'Q8429': 7, 'Q979079': 6, 'Q1151279': 3}\n",
      "8 september {'Q2850': 1588}\n",
      "8½ {'Q12018': 22}\n",
      "zware aardbeving {'Q214866': 2, 'Q43777': 15, 'Q151835': 2, 'Q211386': 3, 'Q1798567': 2}\n",
      "aardbeving {'Q214866': 3, 'Q1903316': 3, 'Q749610': 3, 'Q1053476': 2, 'Q116688': 8, 'Q16068825': 2, 'Q191055': 5, 'Q328663': 7, 'Q210137': 2, 'Q2441400': 7, 'Q43777': 9, 'Q1909872': 2, 'Q19830062': 5, 'Q2885880': 2, 'Q3245204': 3, 'Q274498': 2, 'Q27850765': 2, 'Q2483366': 2, 'Q3161000': 2, 'Q38821': 3, 'Q939106': 6, 'Q1082836': 3, 'Q130754': 2, 'Q1847006': 2, 'Q1348954': 3, 'Q26689701': 3, 'Q745411': 3, 'Q151835': 4, 'Q1990627': 2, 'Q1348910': 5, 'Q211386': 7, 'Q7944': 1417, 'Q867691': 5, 'Q152033': 4, 'Q47144124': 2, 'Q207918': 5, 'Q1798567': 3}\n",
      "8½ {'Q12018': 22}\n",
      "mexico {'Q588480': 2, 'Q260965': 6, 'Q15296784': 18, 'Q599923': 19, 'Q6797304': 4, 'Q285658': 4, 'Q170603': 3, 'Q55712': 11, 'Q641583': 3, 'Q50647': 3, 'Q96': 10551, 'Q2151952': 12, 'Q941143': 2, 'Q8109': 2, 'Q48311380': 2, 'Q108810124': 4, 'Q51636908': 3, 'Q1489': 105, 'Q478399': 6, 'Q82112': 263, 'Q371334': 16, 'Q2294975': 9, 'Q764690': 8, 'Q2776455': 6, 'Q29066592': 3, 'Q603181': 3, 'Q173099': 2, 'Q1182352': 11, 'Q1166178': 3, 'Q1926260': 10, 'Q17041307': 5, 'Q150644': 2, 'Q2342098': 7, 'Q2712776': 3, 'Q12123144': 3, 'Q22696198': 2, 'Q164089': 908, 'Q1160682': 22, 'Q1907023': 4, 'Q3098723': 4, 'Q279769': 2, 'Q1041675': 2, 'Q387588': 3, 'Q19826175': 13, 'Q178652': 12, 'Q2376913': 2, 'Q43196466': 9, 'Q1414354': 3, 'Q8429': 7, 'Q979079': 6, 'Q1151279': 3}\n",
      "mexico {'Q588480': 2, 'Q260965': 6, 'Q15296784': 18, 'Q599923': 19, 'Q6797304': 4, 'Q285658': 4, 'Q170603': 3, 'Q55712': 11, 'Q641583': 3, 'Q50647': 3, 'Q96': 10551, 'Q2151952': 12, 'Q941143': 2, 'Q8109': 2, 'Q48311380': 2, 'Q108810124': 4, 'Q51636908': 3, 'Q1489': 105, 'Q478399': 6, 'Q82112': 263, 'Q371334': 16, 'Q2294975': 9, 'Q764690': 8, 'Q2776455': 6, 'Q29066592': 3, 'Q603181': 3, 'Q173099': 2, 'Q1182352': 11, 'Q1166178': 3, 'Q1926260': 10, 'Q17041307': 5, 'Q150644': 2, 'Q2342098': 7, 'Q2712776': 3, 'Q12123144': 3, 'Q22696198': 2, 'Q164089': 908, 'Q1160682': 22, 'Q1907023': 4, 'Q3098723': 4, 'Q279769': 2, 'Q1041675': 2, 'Q387588': 3, 'Q19826175': 13, 'Q178652': 12, 'Q2376913': 2, 'Q43196466': 9, 'Q1414354': 3, 'Q8429': 7, 'Q979079': 6, 'Q1151279': 3}\n",
      "middernacht {'Q36402': 104}\n",
      "plaatselijke tijd {'Q6940': 2}\n",
      "aardbeving {'Q214866': 3, 'Q1903316': 3, 'Q749610': 3, 'Q1053476': 2, 'Q116688': 8, 'Q16068825': 2, 'Q191055': 5, 'Q328663': 7, 'Q210137': 2, 'Q2441400': 7, 'Q43777': 9, 'Q1909872': 2, 'Q19830062': 5, 'Q2885880': 2, 'Q3245204': 3, 'Q274498': 2, 'Q27850765': 2, 'Q2483366': 2, 'Q3161000': 2, 'Q38821': 3, 'Q939106': 6, 'Q1082836': 3, 'Q130754': 2, 'Q1847006': 2, 'Q1348954': 3, 'Q26689701': 3, 'Q745411': 3, 'Q151835': 4, 'Q1990627': 2, 'Q1348910': 5, 'Q211386': 7, 'Q7944': 1417, 'Q867691': 5, 'Q152033': 4, 'Q47144124': 2, 'Q207918': 5, 'Q1798567': 3}\n",
      "mexicaanse {'Q96': 2011, 'Q581921': 2, 'Q764690': 2, 'Q207965': 4}\n",
      "seismologische {'Q83371': 17}\n",
      "beving {'Q1903316': 25}\n",
      "8½ {'Q12018': 22}\n",
      "de schaal van richter {'Q38768': 6}\n",
      "schaal van richter {'Q38768': 719}\n",
      "richter {'Q2777659': 54, 'Q81240': 4, 'Q2874233': 13, 'Q38768': 7, 'Q224831': 3}\n",
      "stille oceaan {'Q184425': 2, 'Q98': 1274}\n",
      "100 kilometer {'Q1847570': 5}\n",
      "oaxaca {'Q13906835': 3, 'Q131429': 71, 'Q34110': 383, 'Q345582': 2}\n",
      "chiapas {'Q12187096': 13, 'Q1426348': 16, 'Q60123': 394}\n",
      "guatemala {'Q1536706': 5, 'Q844729': 3, 'Q695660': 33, 'Q3667955': 3, 'Q270168': 188, 'Q774': 2109}\n",
      "beving {'Q1903316': 25}\n",
      "slachtoffer {'Q1851760': 351, 'Q181600': 6}\n",
      "aardbeving {'Q214866': 3, 'Q1903316': 3, 'Q749610': 3, 'Q1053476': 2, 'Q116688': 8, 'Q16068825': 2, 'Q191055': 5, 'Q328663': 7, 'Q210137': 2, 'Q2441400': 7, 'Q43777': 9, 'Q1909872': 2, 'Q19830062': 5, 'Q2885880': 2, 'Q3245204': 3, 'Q274498': 2, 'Q27850765': 2, 'Q2483366': 2, 'Q3161000': 2, 'Q38821': 3, 'Q939106': 6, 'Q1082836': 3, 'Q130754': 2, 'Q1847006': 2, 'Q1348954': 3, 'Q26689701': 3, 'Q745411': 3, 'Q151835': 4, 'Q1990627': 2, 'Q1348910': 5, 'Q211386': 7, 'Q7944': 1417, 'Q867691': 5, 'Q152033': 4, 'Q47144124': 2, 'Q207918': 5, 'Q1798567': 3}\n",
      "voelen {'Q205555': 14, 'Q328835': 16}\n",
      "mexico {'Q588480': 2, 'Q260965': 6, 'Q15296784': 18, 'Q599923': 19, 'Q6797304': 4, 'Q285658': 4, 'Q170603': 3, 'Q55712': 11, 'Q641583': 3, 'Q50647': 3, 'Q96': 10551, 'Q2151952': 12, 'Q941143': 2, 'Q8109': 2, 'Q48311380': 2, 'Q108810124': 4, 'Q51636908': 3, 'Q1489': 105, 'Q478399': 6, 'Q82112': 263, 'Q371334': 16, 'Q2294975': 9, 'Q764690': 8, 'Q2776455': 6, 'Q29066592': 3, 'Q603181': 3, 'Q173099': 2, 'Q1182352': 11, 'Q1166178': 3, 'Q1926260': 10, 'Q17041307': 5, 'Q150644': 2, 'Q2342098': 7, 'Q2712776': 3, 'Q12123144': 3, 'Q22696198': 2, 'Q164089': 908, 'Q1160682': 22, 'Q1907023': 4, 'Q3098723': 4, 'Q279769': 2, 'Q1041675': 2, 'Q387588': 3, 'Q19826175': 13, 'Q178652': 12, 'Q2376913': 2, 'Q43196466': 9, 'Q1414354': 3, 'Q8429': 7, 'Q979079': 6, 'Q1151279': 3}\n",
      "mexico stad {'Q1489': 23}\n",
      "mexico-stad {'Q1489': 2470, 'Q665894': 3, 'Q927513': 2}\n",
      "paniek {'Q696490': 3, 'Q208450': 45, 'Q2788113': 5}\n",
      "aardbeving {'Q214866': 3, 'Q1903316': 3, 'Q749610': 3, 'Q1053476': 2, 'Q116688': 8, 'Q16068825': 2, 'Q191055': 5, 'Q328663': 7, 'Q210137': 2, 'Q2441400': 7, 'Q43777': 9, 'Q1909872': 2, 'Q19830062': 5, 'Q2885880': 2, 'Q3245204': 3, 'Q274498': 2, 'Q27850765': 2, 'Q2483366': 2, 'Q3161000': 2, 'Q38821': 3, 'Q939106': 6, 'Q1082836': 3, 'Q130754': 2, 'Q1847006': 2, 'Q1348954': 3, 'Q26689701': 3, 'Q745411': 3, 'Q151835': 4, 'Q1990627': 2, 'Q1348910': 5, 'Q211386': 7, 'Q7944': 1417, 'Q867691': 5, 'Q152033': 4, 'Q47144124': 2, 'Q207918': 5, 'Q1798567': 3}\n",
      "stroomuitval {'Q828827': 45}\n",
      "beving {'Q1903316': 25}\n",
      "tsunami {'Q8070': 465, 'Q36204': 6, 'Q2169285': 6, 'Q1211774': 4, 'Q130754': 22, 'Q15042655': 21}\n",
      "aardbeving {'Q214866': 3, 'Q1903316': 3, 'Q749610': 3, 'Q1053476': 2, 'Q116688': 8, 'Q16068825': 2, 'Q191055': 5, 'Q328663': 7, 'Q210137': 2, 'Q2441400': 7, 'Q43777': 9, 'Q1909872': 2, 'Q19830062': 5, 'Q2885880': 2, 'Q3245204': 3, 'Q274498': 2, 'Q27850765': 2, 'Q2483366': 2, 'Q3161000': 2, 'Q38821': 3, 'Q939106': 6, 'Q1082836': 3, 'Q130754': 2, 'Q1847006': 2, 'Q1348954': 3, 'Q26689701': 3, 'Q745411': 3, 'Q151835': 4, 'Q1990627': 2, 'Q1348910': 5, 'Q211386': 7, 'Q7944': 1417, 'Q867691': 5, 'Q152033': 4, 'Q47144124': 2, 'Q207918': 5, 'Q1798567': 3}\n",
      "kaliber {'Q170417': 393, 'Q3812588': 2, 'Q256690': 14}\n",
      "chili {'Q147131': 2, 'Q863794': 2, 'Q683744': 2, 'Q165199': 3, 'Q1896262': 11, 'Q589856': 7, 'Q143856': 2, 'Q2342527': 2, 'Q172025': 757, 'Q2105105': 10, 'Q393170': 2, 'Q602656': 2, 'Q261405': 7, 'Q298': 4000, 'Q1072889': 2, 'Q755107': 2, 'Q15293622': 12, 'Q606832': 9}\n",
      "zware aardbeving {'Q214866': 2, 'Q43777': 15, 'Q151835': 2, 'Q211386': 3, 'Q1798567': 2}\n",
      "aardbeving {'Q214866': 3, 'Q1903316': 3, 'Q749610': 3, 'Q1053476': 2, 'Q116688': 8, 'Q16068825': 2, 'Q191055': 5, 'Q328663': 7, 'Q210137': 2, 'Q2441400': 7, 'Q43777': 9, 'Q1909872': 2, 'Q19830062': 5, 'Q2885880': 2, 'Q3245204': 3, 'Q274498': 2, 'Q27850765': 2, 'Q2483366': 2, 'Q3161000': 2, 'Q38821': 3, 'Q939106': 6, 'Q1082836': 3, 'Q130754': 2, 'Q1847006': 2, 'Q1348954': 3, 'Q26689701': 3, 'Q745411': 3, 'Q151835': 4, 'Q1990627': 2, 'Q1348910': 5, 'Q211386': 7, 'Q7944': 1417, 'Q867691': 5, 'Q152033': 4, 'Q47144124': 2, 'Q207918': 5, 'Q1798567': 3}\n",
      "mexico {'Q588480': 2, 'Q260965': 6, 'Q15296784': 18, 'Q599923': 19, 'Q6797304': 4, 'Q285658': 4, 'Q170603': 3, 'Q55712': 11, 'Q641583': 3, 'Q50647': 3, 'Q96': 10551, 'Q2151952': 12, 'Q941143': 2, 'Q8109': 2, 'Q48311380': 2, 'Q108810124': 4, 'Q51636908': 3, 'Q1489': 105, 'Q478399': 6, 'Q82112': 263, 'Q371334': 16, 'Q2294975': 9, 'Q764690': 8, 'Q2776455': 6, 'Q29066592': 3, 'Q603181': 3, 'Q173099': 2, 'Q1182352': 11, 'Q1166178': 3, 'Q1926260': 10, 'Q17041307': 5, 'Q150644': 2, 'Q2342098': 7, 'Q2712776': 3, 'Q12123144': 3, 'Q22696198': 2, 'Q164089': 908, 'Q1160682': 22, 'Q1907023': 4, 'Q3098723': 4, 'Q279769': 2, 'Q1041675': 2, 'Q387588': 3, 'Q19826175': 13, 'Q178652': 12, 'Q2376913': 2, 'Q43196466': 9, 'Q1414354': 3, 'Q8429': 7, 'Q979079': 6, 'Q1151279': 3}\n",
      "beving {'Q1903316': 25}\n",
      "nieto {'Q251846': 86}\n",
      "mexico {'Q588480': 2, 'Q260965': 6, 'Q15296784': 18, 'Q599923': 19, 'Q6797304': 4, 'Q285658': 4, 'Q170603': 3, 'Q55712': 11, 'Q641583': 3, 'Q50647': 3, 'Q96': 10551, 'Q2151952': 12, 'Q941143': 2, 'Q8109': 2, 'Q48311380': 2, 'Q108810124': 4, 'Q51636908': 3, 'Q1489': 105, 'Q478399': 6, 'Q82112': 263, 'Q371334': 16, 'Q2294975': 9, 'Q764690': 8, 'Q2776455': 6, 'Q29066592': 3, 'Q603181': 3, 'Q173099': 2, 'Q1182352': 11, 'Q1166178': 3, 'Q1926260': 10, 'Q17041307': 5, 'Q150644': 2, 'Q2342098': 7, 'Q2712776': 3, 'Q12123144': 3, 'Q22696198': 2, 'Q164089': 908, 'Q1160682': 22, 'Q1907023': 4, 'Q3098723': 4, 'Q279769': 2, 'Q1041675': 2, 'Q387588': 3, 'Q19826175': 13, 'Q178652': 12, 'Q2376913': 2, 'Q43196466': 9, 'Q1414354': 3, 'Q8429': 7, 'Q979079': 6, 'Q1151279': 3}\n",
      "mexicaans {'Q82112': 2, 'Q96': 2266}\n",
      "enrique {'Q552058': 12, 'Q47122': 2, 'Q1803019': 2, 'Q2290535': 2}\n",
      "enrique peña {'Q296741': 5}\n",
      "enrique peña nieto {'Q296741': 36}\n",
      "peña {'Q3313041': 8, 'Q417545': 4, 'Q2497712': 6, 'Q6125576': 13, 'Q532103': 12}\n",
      "nieto {'Q251846': 86}\n",
      "dagen van nationale rouw {'Q1127742': 16}\n",
      "nationale rouw {'Q13878715': 2, 'Q1127742': 18}\n",
      "rouw {'Q2450206': 2, 'Q750652': 157}\n",
      "afgekondigd {'Q446780': 2}\n",
      "dodental {'Q1435473': 3}\n"
     ]
    }
   ],
   "source": [
    "def get_matches(surface_trie, text, language=None):\n",
    "    for normtext in normalize(text, language=language):\n",
    "        normtoks = vw_tok(normtext)\n",
    "        for i, tok in enumerate(normtoks):\n",
    "            for comp in surface_trie.keys(tok):\n",
    "                comp_toks = vw_tok(comp)\n",
    "                if normtoks[i : i + len(comp_toks)] == comp_toks:\n",
    "                    yield comp\n",
    "\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "data = pd.read_csv(\"evaluation/Mewsli-9/nl.tsv\", sep=\"\\t\", nrows=10, header=None)\n",
    "data[1] = data[1].map(json.loads)\n",
    "_, mention_ent, text = data.iloc[6]\n",
    "print(text)\n",
    "print(mention_ent)\n",
    "\n",
    "\n",
    "for match in get_matches(surface_trie, text):\n",
    "    # NIL match\n",
    "    weights = surface_weights[match]\n",
    "    print(match, weights)\n",
    "    # for label in vw_label_lines(weights, -1, comp, ent_feats):\n",
    "    #     labels.append(label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "19781988-b9e6-48ea-ac9e-743b36e23a1e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "41bc8360-80c2-45a2-a160-31e86f621562",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "wc            1.0\n",
       "aec          59.0\n",
       "ratio         0.0\n",
       "ev_count      2.0\n",
       "ev_linked     0.0\n",
       "Name: op de rails, dtype: Float64"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.loc[\"op de rails\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3e8221c2-e8e1-4633-8891-216c1080252a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['shared |s ik ging naar amsterdam en rotterdam',\n",
       " '9899:0 |l amsterdam=9899 ',\n",
       " '727:1 |l amsterdam=727 ',\n",
       " '478771:1 |l amsterdam=478771 ',\n",
       " '50719:1 |l amsterdam=50719 ',\n",
       " '2060132:1 |l amsterdam=2060132 ',\n",
       " '214341:1 |l amsterdam=214341 ',\n",
       " '194215:1 |l amsterdam=194215 ',\n",
       " '26674641:1 |l amsterdam=26674641 ',\n",
       " '1127380:1 |l amsterdam=1127380 ',\n",
       " '9694:1 |l amsterdam=9694 ',\n",
       " '2691069:1 |l amsterdam=2691069 ',\n",
       " '2393888:1 |l amsterdam=2393888 ',\n",
       " '478785:1 |l amsterdam=478785 ',\n",
       " '1397383:1 |l amsterdam=1397383 ',\n",
       " '122781:1 |l amsterdam=122781 ',\n",
       " '959016:1 |l amsterdam=959016 ',\n",
       " '5715626:1 |l amsterdam=5715626 ',\n",
       " '2060246:1 |l amsterdam=2060246 ',\n",
       " '4748823:1 |l amsterdam=4748823 ',\n",
       " '2049529:1 |l amsterdam=2049529 ',\n",
       " '95630476:1 |l amsterdam=95630476 ',\n",
       " '683829:1 |l amsterdam=683829 ',\n",
       " '478456:1 |l amsterdam=478456 ',\n",
       " '4549041:1 |l amsterdam=4549041 ',\n",
       " '13427674:1 |l amsterdam=13427674 ',\n",
       " '585429:1 |l amsterdam=585429 ',\n",
       " '505639:1 |l amsterdam=505639 ',\n",
       " '2276292:1 |l amsterdam=2276292 ',\n",
       " '-1:0 |l ik=-1 ',\n",
       " '49404:1 |l ik=49404 ',\n",
       " '2422744:1 |l ik=2422744 ',\n",
       " '1143089:1 |l ik=1143089 ',\n",
       " '-1:0 |l en=-1 ',\n",
       " '1860:1 |l en=1860 ',\n",
       " '1377447:1 |l en=1377447 ',\n",
       " '191081:1 |l en=191081 ',\n",
       " '933:1 |l en=933 ',\n",
       " '-1:0 |l rotterdam=-1 ',\n",
       " '34370:1 |l rotterdam=34370 ',\n",
       " '2680952:1 |l rotterdam=2680952 ',\n",
       " '201284:1 |l rotterdam=201284 ',\n",
       " '801388:1 |l rotterdam=801388 ',\n",
       " '1027807:1 |l rotterdam=1027807 ',\n",
       " '166127:1 |l rotterdam=166127 ',\n",
       " '1429091:1 |l rotterdam=1429091 ',\n",
       " '783945:1 |l rotterdam=783945 ',\n",
       " '2329167:1 |l rotterdam=2329167 ',\n",
       " '931364:1 |l rotterdam=931364 ',\n",
       " '1341108:1 |l rotterdam=1341108 ',\n",
       " '299701:1 |l rotterdam=299701 ',\n",
       " '2166239:1 |l rotterdam=2166239 ',\n",
       " '2856391:1 |l rotterdam=2856391 ',\n",
       " '656807:1 |l rotterdam=656807 ',\n",
       " '633529:1 |l rotterdam=633529 ',\n",
       " '849230:1 |l rotterdam=849230 ',\n",
       " '1774596:1 |l rotterdam=1774596 ',\n",
       " '2930113:1 |l rotterdam=2930113 ',\n",
       " '1866458:1 |l rotterdam=1866458 ',\n",
       " '',\n",
       " '']"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from minimel.vectorize import vw\n",
    "import pathlib\n",
    "\n",
    "fname = \"wiki/nlwiki-20220301/experiments/clean-q0.25.json\"\n",
    "vw(\n",
    "    ['id\\t{\"amsterdam\": 9899}\\tik ging naar amsterdam en Rotterdam'],\n",
    "    pathlib.Path(fname),\n",
    ")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:minimel]",
   "language": "python",
   "name": "conda-env-minimel-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.17"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}