DigCon_Algorithms.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "54897330-11bc-402a-a4dc-6ecba774999a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done!\n"
     ]
    }
   ],
   "source": [
    "# Analysis terms: [ai/art. intell., machine learning, algorithm]\n",
    "\n",
    "project = \"algorithm\"\n",
    "\n",
    "import pandas\n",
    "\n",
    "from web_observatory import start_project\n",
    "start_project(project)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "13e4e849-5b36-47a3-b950-737f6966d56e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'query': 'conservation (algorithms OR algorithmic) has:links', 'tweet.fields': 'created_at,entities,public_metrics', 'max_results': 100, 'start_time': '2006-03-21T00:00:00.00Z', 'end_time': '2022-01-01T00:00:00.00Z'}\n",
      "Done!\n"
     ]
    }
   ],
   "source": [
    "# Get tweets about AI/ML in nature conservation (Historical focus)\n",
    "from web_observatory import search_twitter\n",
    "algo_twitter = search_twitter(q = 'conservation (algorithms OR algorithmic) has:links', project=project)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ef98ca88-4658-4447-92a0-73615dc105db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done\n"
     ]
    }
   ],
   "source": [
    "# Get Google results about AI/ML in nature conservation (Relevance)\n",
    "from web_observatory import search_google\n",
    "algo_google = search_google(q = 'nature conservation algorithm', project = project)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5bdeec06-82c8-4c9d-958e-c15df0e58984",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compile results\n",
    "from web_observatory import twitter_process \n",
    "twitter = twitter_process(project=project)\n",
    "from web_observatory import google_process\n",
    "google = google_process(datatype = \"CSV\", project=project)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f0ef2456-3f1f-436c-8f74-91f40e64ad6c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>link</th>\n",
       "      <th>metrics</th>\n",
       "      <th>source</th>\n",
       "      <th>query</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>domain</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>wiley</th>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sagepub</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nih</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>amazon</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>springer</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>researchgate</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sciencedirect</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>scitation</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>scitechdaily</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>siam</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>smartparks</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>springeropen</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>psu</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tandfonline</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>theguardian</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>thehill</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ucl</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ucsc</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>uga</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>usc</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               date  link  metrics  source  query\n",
       "domain                                           \n",
       "wiley             0     5        0       5      5\n",
       "sagepub           0     2        0       2      2\n",
       "nih               0     2        0       2      2\n",
       "amazon            0     1        0       1      1\n",
       "springer          0     1        0       1      1\n",
       "researchgate      0     1        0       1      1\n",
       "sciencedirect     0     1        0       1      1\n",
       "scitation         0     1        0       1      1\n",
       "scitechdaily      0     1        0       1      1\n",
       "siam              0     1        0       1      1\n",
       "smartparks        0     1        0       1      1\n",
       "springeropen      0     1        0       1      1\n",
       "psu               0     1        0       1      1\n",
       "tandfonline       0     1        0       1      1\n",
       "theguardian       0     1        0       1      1\n",
       "thehill           0     1        0       1      1\n",
       "ucl               0     1        0       1      1\n",
       "ucsc              0     1        0       1      1\n",
       "uga               0     1        0       1      1\n",
       "usc               0     1        0       1      1"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Extract organizations from Twitter/Google results\n",
    "from web_observatory import get_domains\n",
    "twitter = get_domains(twitter)\n",
    "google = get_domains(google)\n",
    "twitter.groupby(by=\"domain\").count().sort_values(by=\"link\", ascending=False).head(20)\n",
    "google.groupby(by=\"domain\").count().sort_values(by=\"link\", ascending=False).head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2a31952a-9dae-46d4-b0bd-8e255bcb2367",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Process all data together \n",
    "from web_observatory import search_merge\n",
    "pages = search_merge(twitter = twitter, google = google, project = project)\n",
    "pages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "b90dae61-8bcb-441b-b96d-a14295d1e63a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>link</th>\n",
       "      <th>metrics</th>\n",
       "      <th>source</th>\n",
       "      <th>query</th>\n",
       "      <th>domain</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2011-08-27T05:07:41.000Z</td>\n",
       "      <td>http://amzn.to/qOpv8k</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>conservation (algorithms OR algorithmic) has:l...</td>\n",
       "      <td>amzn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2011-11-02T03:23:01.000Z</td>\n",
       "      <td>http://lnk.co/HL2EX</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>conservation (algorithms OR algorithmic) has:l...</td>\n",
       "      <td>lnk</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2011-11-05T10:07:25.000Z</td>\n",
       "      <td>http://amzn.to/ualRkq</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>conservation (algorithms OR algorithmic) has:l...</td>\n",
       "      <td>amzn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2011-11-24T05:05:58.000Z</td>\n",
       "      <td>http://bit.ly/vLUoyA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>conservation (algorithms OR algorithmic) has:l...</td>\n",
       "      <td>bit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2011-12-22T08:46:02.000Z</td>\n",
       "      <td>http://amzn.to/t8LXDS</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>conservation (algorithms OR algorithmic) has:l...</td>\n",
       "      <td>amzn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>NaN</td>\n",
       "      <td>https://cran.r-project.org/web/packages/priori...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Google</td>\n",
       "      <td>nature conservation algorithm</td>\n",
       "      <td>r-project</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>NaN</td>\n",
       "      <td>https://www.mdpi.com/1424-8220/19/3/553</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Google</td>\n",
       "      <td>nature conservation algorithm</td>\n",
       "      <td>mdpi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>NaN</td>\n",
       "      <td>http://files.eric.ed.gov/fulltext/ED193078.pdf</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Google</td>\n",
       "      <td>nature conservation algorithm</td>\n",
       "      <td>ed</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>NaN</td>\n",
       "      <td>https://www.nrcs.usda.gov/wps/portal/nrcs/main...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Google</td>\n",
       "      <td>nature conservation algorithm</td>\n",
       "      <td>usda</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>NaN</td>\n",
       "      <td>https://jwcn-eurasipjournals.springeropen.com/...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Google</td>\n",
       "      <td>nature conservation algorithm</td>\n",
       "      <td>springeropen</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>237 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                        date  \\\n",
       "0   2011-08-27T05:07:41.000Z   \n",
       "1   2011-11-02T03:23:01.000Z   \n",
       "2   2011-11-05T10:07:25.000Z   \n",
       "3   2011-11-24T05:05:58.000Z   \n",
       "4   2011-12-22T08:46:02.000Z   \n",
       "..                       ...   \n",
       "42                       NaN   \n",
       "43                       NaN   \n",
       "44                       NaN   \n",
       "47                       NaN   \n",
       "48                       NaN   \n",
       "\n",
       "                                                 link  metrics   source  \\\n",
       "0                               http://amzn.to/qOpv8k      0.0  Twitter   \n",
       "1                                 http://lnk.co/HL2EX      0.0  Twitter   \n",
       "2                               http://amzn.to/ualRkq      0.0  Twitter   \n",
       "3                                http://bit.ly/vLUoyA      0.0  Twitter   \n",
       "4                               http://amzn.to/t8LXDS      0.0  Twitter   \n",
       "..                                                ...      ...      ...   \n",
       "42  https://cran.r-project.org/web/packages/priori...      NaN   Google   \n",
       "43            https://www.mdpi.com/1424-8220/19/3/553      NaN   Google   \n",
       "44     http://files.eric.ed.gov/fulltext/ED193078.pdf      NaN   Google   \n",
       "47  https://www.nrcs.usda.gov/wps/portal/nrcs/main...      NaN   Google   \n",
       "48  https://jwcn-eurasipjournals.springeropen.com/...      NaN   Google   \n",
       "\n",
       "                                                query        domain  \n",
       "0   conservation (algorithms OR algorithmic) has:l...          amzn  \n",
       "1   conservation (algorithms OR algorithmic) has:l...           lnk  \n",
       "2   conservation (algorithms OR algorithmic) has:l...          amzn  \n",
       "3   conservation (algorithms OR algorithmic) has:l...           bit  \n",
       "4   conservation (algorithms OR algorithmic) has:l...          amzn  \n",
       "..                                                ...           ...  \n",
       "42                      nature conservation algorithm     r-project  \n",
       "43                      nature conservation algorithm          mdpi  \n",
       "44                      nature conservation algorithm            ed  \n",
       "47                      nature conservation algorithm          usda  \n",
       "48                      nature conservation algorithm  springeropen  \n",
       "\n",
       "[237 rows x 6 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Manage duplication of links here....\n",
    "## Load previous \"all_search_results\"\n",
    "previous = pandas.read_csv(\"AI_ML/all_search_results.csv\")\n",
    "## De-duplicate from this list of pages where there is a match on previous search\n",
    "pages = pages[~pages[\"link\"].isin(previous[\"link\"])]\n",
    "pages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "fc992c28-427b-46d5-88d4-f342c5b3c88b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/enost/Downloads/observatory_dev_v02.py:693: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  pages[\"full_url\"] = None # The actual url and not some bit.ly url\n",
      "/Users/enost/Downloads/observatory_dev_v02.py:694: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  pages[\"url_meta\"] = None # Possible additional information\n",
      "/Users/enost/Downloads/observatory_dev_v02.py:695: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  pages[\"wm_url\"] = None # We may have this when we have previously saved pages to the WM\n",
      "/Users/enost/Downloads/observatory_dev_v02.py:696: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  pages[\"wm_meta\"] = None # For storing information about the WM snapshot that was available\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236\r"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>date</th>\n",
       "      <th>link</th>\n",
       "      <th>metrics</th>\n",
       "      <th>source</th>\n",
       "      <th>query</th>\n",
       "      <th>domain</th>\n",
       "      <th>full_url</th>\n",
       "      <th>url_meta</th>\n",
       "      <th>wm_url</th>\n",
       "      <th>wm_meta</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2011-08-27T05:07:41.000Z</td>\n",
       "      <td>http://amzn.to/qOpv8k</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>conservation (algorithms OR algorithmic) has:l...</td>\n",
       "      <td>amzn</td>\n",
       "      <td>https://www.amazon.com/Symmetries-Partial-Diff...</td>\n",
       "      <td>None</td>\n",
       "      <td>http://web.archive.org/web/20130704004148id_/h...</td>\n",
       "      <td>No Time-of-Tweet Versions; Got Older Version</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2011-11-02T03:23:01.000Z</td>\n",
       "      <td>http://lnk.co/HL2EX</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>conservation (algorithms OR algorithmic) has:l...</td>\n",
       "      <td>lnk</td>\n",
       "      <td>http://lnk.co/HL2EX</td>\n",
       "      <td>None</td>\n",
       "      <td>http://web.archive.org/web/20110915112900id_/h...</td>\n",
       "      <td>Time-of-Tweet Version</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2011-11-05T10:07:25.000Z</td>\n",
       "      <td>http://amzn.to/ualRkq</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>conservation (algorithms OR algorithmic) has:l...</td>\n",
       "      <td>amzn</td>\n",
       "      <td>https://www.amazon.com/Symmetries-Partial-Diff...</td>\n",
       "      <td>None</td>\n",
       "      <td>http://web.archive.org/web/20130704004148id_/h...</td>\n",
       "      <td>No Time-of-Tweet Versions; Got Older Version</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>2011-11-24T05:05:58.000Z</td>\n",
       "      <td>http://bit.ly/vLUoyA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>conservation (algorithms OR algorithmic) has:l...</td>\n",
       "      <td>bit</td>\n",
       "      <td>https://www.freelancer.com/projects/PHP-Javasc...</td>\n",
       "      <td>None</td>\n",
       "      <td>http://web.archive.org/web/20111228103051id_/h...</td>\n",
       "      <td>No Time-of-Tweet Versions; Got Older Version</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>2011-12-22T08:46:02.000Z</td>\n",
       "      <td>http://amzn.to/t8LXDS</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>conservation (algorithms OR algorithmic) has:l...</td>\n",
       "      <td>amzn</td>\n",
       "      <td>https://www.amazon.com/Symmetries-Partial-Diff...</td>\n",
       "      <td>None</td>\n",
       "      <td>http://web.archive.org/web/20130704004148id_/h...</td>\n",
       "      <td>No Time-of-Tweet Versions; Got Older Version</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>232</th>\n",
       "      <td>42</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://cran.r-project.org/web/packages/priori...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Google</td>\n",
       "      <td>nature conservation algorithm</td>\n",
       "      <td>r-project</td>\n",
       "      <td>https://cran.r-project.org/web/packages/priori...</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>Not historical</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>233</th>\n",
       "      <td>43</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://www.mdpi.com/1424-8220/19/3/553</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Google</td>\n",
       "      <td>nature conservation algorithm</td>\n",
       "      <td>mdpi</td>\n",
       "      <td>https://www.mdpi.com/1424-8220/19/3/553</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>Not historical</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>234</th>\n",
       "      <td>44</td>\n",
       "      <td>NaN</td>\n",
       "      <td>http://files.eric.ed.gov/fulltext/ED193078.pdf</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Google</td>\n",
       "      <td>nature conservation algorithm</td>\n",
       "      <td>ed</td>\n",
       "      <td>https://files.eric.ed.gov/fulltext/ED193078.pdf</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>Not historical</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>235</th>\n",
       "      <td>47</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://www.nrcs.usda.gov/wps/portal/nrcs/main...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Google</td>\n",
       "      <td>nature conservation algorithm</td>\n",
       "      <td>usda</td>\n",
       "      <td>https://www.nrcs.usda.gov/wps/portal/nrcs/main...</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>Not historical</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>236</th>\n",
       "      <td>48</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://jwcn-eurasipjournals.springeropen.com/...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Google</td>\n",
       "      <td>nature conservation algorithm</td>\n",
       "      <td>springeropen</td>\n",
       "      <td>https://jwcn-eurasipjournals.springeropen.com/...</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>Not historical</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>237 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     index                      date  \\\n",
       "0        0  2011-08-27T05:07:41.000Z   \n",
       "1        1  2011-11-02T03:23:01.000Z   \n",
       "2        2  2011-11-05T10:07:25.000Z   \n",
       "3        3  2011-11-24T05:05:58.000Z   \n",
       "4        4  2011-12-22T08:46:02.000Z   \n",
       "..     ...                       ...   \n",
       "232     42                       NaN   \n",
       "233     43                       NaN   \n",
       "234     44                       NaN   \n",
       "235     47                       NaN   \n",
       "236     48                       NaN   \n",
       "\n",
       "                                                  link  metrics   source  \\\n",
       "0                                http://amzn.to/qOpv8k      0.0  Twitter   \n",
       "1                                  http://lnk.co/HL2EX      0.0  Twitter   \n",
       "2                                http://amzn.to/ualRkq      0.0  Twitter   \n",
       "3                                 http://bit.ly/vLUoyA      0.0  Twitter   \n",
       "4                                http://amzn.to/t8LXDS      0.0  Twitter   \n",
       "..                                                 ...      ...      ...   \n",
       "232  https://cran.r-project.org/web/packages/priori...      NaN   Google   \n",
       "233            https://www.mdpi.com/1424-8220/19/3/553      NaN   Google   \n",
       "234     http://files.eric.ed.gov/fulltext/ED193078.pdf      NaN   Google   \n",
       "235  https://www.nrcs.usda.gov/wps/portal/nrcs/main...      NaN   Google   \n",
       "236  https://jwcn-eurasipjournals.springeropen.com/...      NaN   Google   \n",
       "\n",
       "                                                 query        domain  \\\n",
       "0    conservation (algorithms OR algorithmic) has:l...          amzn   \n",
       "1    conservation (algorithms OR algorithmic) has:l...           lnk   \n",
       "2    conservation (algorithms OR algorithmic) has:l...          amzn   \n",
       "3    conservation (algorithms OR algorithmic) has:l...           bit   \n",
       "4    conservation (algorithms OR algorithmic) has:l...          amzn   \n",
       "..                                                 ...           ...   \n",
       "232                      nature conservation algorithm     r-project   \n",
       "233                      nature conservation algorithm          mdpi   \n",
       "234                      nature conservation algorithm            ed   \n",
       "235                      nature conservation algorithm          usda   \n",
       "236                      nature conservation algorithm  springeropen   \n",
       "\n",
       "                                              full_url url_meta  \\\n",
       "0    https://www.amazon.com/Symmetries-Partial-Diff...     None   \n",
       "1                                  http://lnk.co/HL2EX     None   \n",
       "2    https://www.amazon.com/Symmetries-Partial-Diff...     None   \n",
       "3    https://www.freelancer.com/projects/PHP-Javasc...     None   \n",
       "4    https://www.amazon.com/Symmetries-Partial-Diff...     None   \n",
       "..                                                 ...      ...   \n",
       "232  https://cran.r-project.org/web/packages/priori...     None   \n",
       "233            https://www.mdpi.com/1424-8220/19/3/553     None   \n",
       "234    https://files.eric.ed.gov/fulltext/ED193078.pdf     None   \n",
       "235  https://www.nrcs.usda.gov/wps/portal/nrcs/main...     None   \n",
       "236  https://jwcn-eurasipjournals.springeropen.com/...     None   \n",
       "\n",
       "                                                wm_url  \\\n",
       "0    http://web.archive.org/web/20130704004148id_/h...   \n",
       "1    http://web.archive.org/web/20110915112900id_/h...   \n",
       "2    http://web.archive.org/web/20130704004148id_/h...   \n",
       "3    http://web.archive.org/web/20111228103051id_/h...   \n",
       "4    http://web.archive.org/web/20130704004148id_/h...   \n",
       "..                                                 ...   \n",
       "232                                               None   \n",
       "233                                               None   \n",
       "234                                               None   \n",
       "235                                               None   \n",
       "236                                               None   \n",
       "\n",
       "                                          wm_meta  \n",
       "0    No Time-of-Tweet Versions; Got Older Version  \n",
       "1                           Time-of-Tweet Version  \n",
       "2    No Time-of-Tweet Versions; Got Older Version  \n",
       "3    No Time-of-Tweet Versions; Got Older Version  \n",
       "4    No Time-of-Tweet Versions; Got Older Version  \n",
       "..                                            ...  \n",
       "232                                Not historical  \n",
       "233                                Not historical  \n",
       "234                                Not historical  \n",
       "235                                Not historical  \n",
       "236                                Not historical  \n",
       "\n",
       "[237 rows x 11 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get Twitter archives\n",
    "#pages = pandas.read_csv(\"algorithm/all_search_results.csv\")\n",
    "from web_observatory import get_versions\n",
    "results = get_versions(pages, project = project)\n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "401fdf37-0a0b-417c-ab1d-967404c2df18",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done!\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/enost/opt/anaconda3/lib/python3.9/site-packages/pandas/core/indexing.py:1681: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self.obj[key] = empty_value\n",
      "/Users/enost/opt/anaconda3/lib/python3.9/site-packages/pandas/core/indexing.py:1773: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self._setitem_single_column(ilocs[0], value, pi)\n",
      "/Users/enost/Downloads/observatory_dev_v02.py:843: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  versions['domain'] = versions.apply(lambda row: domain(row), axis=1)\n",
      "/Users/enost/Downloads/observatory_dev_v02.py:850: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  versions['date'] = pandas.to_datetime(versions['date']).dt.strftime('%Y-%U') # Convert to year/week\n"
     ]
    }
   ],
   "source": [
    "# Scrape pages\n",
    "## Initialize / De-duplicate full urls\n",
    "#results = pandas.read_csv(project+\"/page_versions_full.csv\")\n",
    "from web_observatory import initialize_scrape\n",
    "urls = initialize_scrape(versions = results, project = project) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "1fa6e5c3-5ed1-4d76-887a-607ee7b5af46",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/enost/Downloads/digcon_scraper/digcon_scraper\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:08 [py.warnings] WARNING: /Users/enost/opt/anaconda3/lib/python3.9/site-packages/scrapy/extensions/feedexport.py:289: ScrapyDeprecationWarning: The `FEED_URI` and `FEED_FORMAT` settings have been deprecated in favor of the `FEEDS` setting. Please see the `FEEDS` setting docs for more details\n",
      "  exporter = cls(crawler)\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inserting: 44\r"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:10 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://onlinelibrary.wiley.com/resolve/doi> (failed 3 times): 503 Service Unavailable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inserting: 86\r"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:17 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://besjournals.onlinelibrary.wiley.com/doi/abs/10.1111/2041-210x.13264> (failed 3 times): 503 Service Unavailable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inserting: 952\r"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:20 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://www.linkedin.com/leap/r> (failed 3 times): 502 Bad Gateway\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inserting: 127\r"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:24 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://www.cell.com/one-earth/fulltext/s2590-3322(21)00292-x> (failed 3 times): 503 Service Unavailable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inserting: 137\r"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:26 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://journals.sagepub.com/doi/abs/10.1177/0309132517740220> (failed 3 times): 503 Service Unavailable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inserting: 164\r"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:27 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://journals.sagepub.com/doi/full/10.1177/0309132517740220> (failed 3 times): 503 Service Unavailable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inserting: 142\r"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:27 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1002/2016wr019987> (failed 3 times): 503 Service Unavailable\n",
      "2022-06-27 13:43:27 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://onlinelibrary.wiley.com/doi/abs/10.1111/gcb.12736> (failed 3 times): 503 Service Unavailable\n",
      "2022-06-27 13:43:27 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://conbio.onlinelibrary.wiley.com/doi/10.1111/cobi.13680> (failed 3 times): 503 Service Unavailable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inserting: 175\r"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:27 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://www.cambridge.org/core/journals/oryx/article/conservation-by-algorithm/45af526d27e654d4fe18a766fddfe9c8> (failed 3 times): 500 Internal Server Error\n",
      "2022-06-27 13:43:27 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://aip.scitation.org/doi/10.1063/1.4931597> (failed 3 times): 503 Service Unavailable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inserting: 158\r"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:28 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://esajournals.onlinelibrary.wiley.com/doi/10.1890/14-1464.1> (failed 3 times): 503 Service Unavailable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inserting: 152\r"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:28 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://www.pnas.org/doi/10.1073/pnas.1813051115> (failed 3 times): 503 Service Unavailable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inserting: 197\r"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:43:30 [scrapy.downloadermiddlewares.retry] ERROR: Gave up retrying <GET https://ieeexplore.ieee.org/document/7438882> (failed 3 times): 503 Service Unavailable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done!ting: 134\r"
     ]
    }
   ],
   "source": [
    "%cd digcon_scraper/digcon_scraper/\n",
    "from web_observatory import scrape\n",
    "scrape(urls)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "c3a64293-1016-4955-add5-f604333a7f91",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/enost/Downloads\n"
     ]
    }
   ],
   "source": [
    "%cd .."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "2a5a0e1b-5059-496c-9777-986e918f74c4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done!\n",
      "Done!\n",
      "Done!\n"
     ]
    }
   ],
   "source": [
    "# Quality control\n",
    "from web_observatory import query\n",
    "## Create a copy of the text table\n",
    "query(\"copy\") # Only need to do this once!\n",
    "## Remove from db where there was no text scraped\n",
    "query(\"filter_empty\")\n",
    "## Remove anything without at least 1 keyword?\n",
    "queryterms = [\"artificial intelligence\", \"machine learning\", \"nature \", \"conservation \", \"algorithm\"] # spaces to avoid certain spam...\n",
    "query(\"filter_words\", terms = queryterms) ## TBD: Need to remove French-language sites?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "102019d2-4e82-4906-bb1f-97032cf8d9ec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>url</th>\n",
       "      <th>domain</th>\n",
       "      <th>source</th>\n",
       "      <th>date</th>\n",
       "      <th>artificial intelligence</th>\n",
       "      <th>ai</th>\n",
       "      <th>machine learning</th>\n",
       "      <th>algorithm</th>\n",
       "      <th>conservation</th>\n",
       "      <th>nature</th>\n",
       "      <th>...</th>\n",
       "      <th>river</th>\n",
       "      <th>arctic</th>\n",
       "      <th>privacy</th>\n",
       "      <th>surveillance</th>\n",
       "      <th>cost</th>\n",
       "      <th>market</th>\n",
       "      <th>decision making</th>\n",
       "      <th>working land</th>\n",
       "      <th>indigenous-led</th>\n",
       "      <th>easement</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>http://web.archive.org/web/20140731104236id_/h...</td>\n",
       "      <td>chesterfieldpost</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>2011-36</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>https://etienne34-blog.tumblr.com/post/1664008...</td>\n",
       "      <td>tumblr</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>2012-04</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>http://web.archive.org/web/20110828190407id_/h...</td>\n",
       "      <td>harriscompanyrec</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>2011-36</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>http://web.archive.org/web/20121114210812id_/h...</td>\n",
       "      <td>webcaviste</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>2011-40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>https://www.freelancer.com/projects/machine-le...</td>\n",
       "      <td>freelancer</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>2012-16</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24805</th>\n",
       "      <td>http://web.archive.org/web/20220430003951id_/h...</td>\n",
       "      <td>biorxiv</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>2019-46</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24806</th>\n",
       "      <td>http://web.archive.org/web/20210517095245id_/h...</td>\n",
       "      <td>wildlabs</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>2019-13</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24807</th>\n",
       "      <td>https://www.miragenews.com/investigating-mathe...</td>\n",
       "      <td>miragenews</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>2021-04</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24808</th>\n",
       "      <td>http://web.archive.org/web/20211226210924id_/h...</td>\n",
       "      <td>nih</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>2021-06</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24809</th>\n",
       "      <td>https://www.mdpi.com/1424-8220/19/3/553</td>\n",
       "      <td>mdpi</td>\n",
       "      <td>Google</td>\n",
       "      <td>None</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>24810 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                     url            domain  \\\n",
       "0      http://web.archive.org/web/20140731104236id_/h...  chesterfieldpost   \n",
       "1      https://etienne34-blog.tumblr.com/post/1664008...            tumblr   \n",
       "2      http://web.archive.org/web/20110828190407id_/h...  harriscompanyrec   \n",
       "3      http://web.archive.org/web/20121114210812id_/h...        webcaviste   \n",
       "4      https://www.freelancer.com/projects/machine-le...        freelancer   \n",
       "...                                                  ...               ...   \n",
       "24805  http://web.archive.org/web/20220430003951id_/h...           biorxiv   \n",
       "24806  http://web.archive.org/web/20210517095245id_/h...          wildlabs   \n",
       "24807  https://www.miragenews.com/investigating-mathe...        miragenews   \n",
       "24808  http://web.archive.org/web/20211226210924id_/h...               nih   \n",
       "24809            https://www.mdpi.com/1424-8220/19/3/553              mdpi   \n",
       "\n",
       "        source     date  artificial intelligence   ai   machine learning  \\\n",
       "0      Twitter  2011-36                        0     0                 0   \n",
       "1      Twitter  2012-04                        0     0                 0   \n",
       "2      Twitter  2011-36                        0     1                 0   \n",
       "3      Twitter  2011-40                        0     0                 0   \n",
       "4      Twitter  2012-16                        0     0                 2   \n",
       "...        ...      ...                      ...   ...               ...   \n",
       "24805  Twitter  2019-46                        0     0                 0   \n",
       "24806  Twitter  2019-13                        2     5                 3   \n",
       "24807  Twitter  2021-04                        0     0                 0   \n",
       "24808  Twitter  2021-06                        0     0                 0   \n",
       "24809   Google     None                        0     0                 0   \n",
       "\n",
       "       algorithm  conservation  nature  ...  river  arctic  privacy  \\\n",
       "0              0             4       0  ...      4       0        1   \n",
       "1              0             1       0  ...      0       0        0   \n",
       "2              0             0       2  ...      0       0        0   \n",
       "3              0             1       0  ...      0       0        0   \n",
       "4              0             6       0  ...      0       0        0   \n",
       "...          ...           ...     ...  ...    ...     ...      ...   \n",
       "24805          4             2       0  ...      0       0        0   \n",
       "24806          3            12       0  ...      0       0        0   \n",
       "24807          1             1       0  ...      0       0        0   \n",
       "24808          1             4       0  ...      0       0        0   \n",
       "24809          3             0       0  ...      0       0        0   \n",
       "\n",
       "       surveillance  cost  market  decision making  working land  \\\n",
       "0                 0     3       1                0             0   \n",
       "1                 0     0       0                0             0   \n",
       "2                 0     0       3                0             0   \n",
       "3                 0     0       0                0             0   \n",
       "4                 0     0       0                0             0   \n",
       "...             ...   ...     ...              ...           ...   \n",
       "24805             0     2       0                0             0   \n",
       "24806             0     0       0                0             0   \n",
       "24807             0     0       0                0             0   \n",
       "24808             0     0       0                0             0   \n",
       "24809             0     3       0                0             0   \n",
       "\n",
       "       indigenous-led  easement  \n",
       "0                   0         0  \n",
       "1                   0         0  \n",
       "2                   0         1  \n",
       "3                   0         0  \n",
       "4                   0         0  \n",
       "...               ...       ...  \n",
       "24805               0         0  \n",
       "24806               0         0  \n",
       "24807               0         0  \n",
       "24808               0         0  \n",
       "24809               0         0  \n",
       "\n",
       "[24810 rows x 25 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Get word counts\n",
    "from web_observatory import query\n",
    "tech_terms = [\"artificial intelligence\", \" ai \", \"machine learning\", \"algorithm\"] # Spaces in ai to capture its use as in AI rather than e.g. said\n",
    "cons_terms = [\"conservation\", \"nature\"]\n",
    "#grammar = [\"is\", \"will\"] # for phrases\n",
    "topics = [\"forest\", \"mountain\", \"wetland\", \"boreal\", \"ocean\", \"river\", \"arctic\"]\n",
    "ethics = [\"privacy\", \"surveillance\", \"cost\"]\n",
    "tools = [\"market\", \"decision making\", \"working land\", \"indigenous-led\", \"easement\"]\n",
    "combined_terms = tech_terms + cons_terms + topics + ethics + tools\n",
    "\n",
    "words = query(\"get_words\", combined_terms)\n",
    "words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "5d97ea18-bd7d-498c-ae2c-b0b094a20e56",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:47:25 [py.warnings] WARNING: /Users/enost/Downloads/observatory_dev_v02.py:1145: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  # Process\n",
      "\n",
      "2022-06-27 13:47:25 [py.warnings] WARNING: /Users/enost/Downloads/observatory_dev_v02.py:1146: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  counts = counts.loc[counts['source'] == \"Twitter\"] # Filter counts to just Twitter sources\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='date'>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='date'>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='date'>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='date'>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Analyze\n",
    "## How has the conversation about AI/ML changed over time? (Twitter frequency/average)\n",
    "from web_observatory import analyze_twitter\n",
    "avg, freq, total = analyze_twitter(words, terms = [\"artificial intelligence\", \" ai \", \"machine learning\", \"algorithm\"])\n",
    "## Where the term conservation is used\n",
    "#, freq, total = analyze_twitter(words.loc[words['nature']>0], [\"artificial intelligence\", \"machine learning\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "ef17a617-10d9-498c-9680-7a9ce4a61745",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 13:48:26 [py.warnings] WARNING: /Users/enost/Downloads/observatory_dev_v02.py:1213: RuntimeWarning: invalid value encountered in long_scalars\n",
      "  for t in terms:\n",
      "\n",
      "2022-06-27 13:48:26 [py.warnings] WARNING: /Users/enost/Downloads/observatory_dev_v02.py:1220: RuntimeWarning: invalid value encountered in long_scalars\n",
      "  m = m[t].count() # Count the number of pages (links) the term was mentioned at least once\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>privacy_avg</th>\n",
       "      <th>surveillance_avg</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>raresites</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>davidsuzuki</th>\n",
       "      <td>10.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grandriver</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wwf</th>\n",
       "      <td>9.43</td>\n",
       "      <td>33.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ecotrust</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>natureconservancy</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  privacy_avg surveillance_avg\n",
       "raresites                 NaN              NaN\n",
       "davidsuzuki              10.0              0.0\n",
       "grandriver                NaN              NaN\n",
       "wwf                      9.43            33.96\n",
       "ecotrust                  0.0              0.0\n",
       "natureconservancy         0.0              0.0"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Are certain organizations (and organization sizes) more likely to mention AI/ML? (Crawl)\n",
    "from web_observatory import analyze_orgs\n",
    "avg, freq = analyze_orgs(words, orgs = [\"raresites\", \"davidsuzuki\", \"grandriver\", \"wwf\", \"ecotrust\", \"natureconservancy\"], terms = [\"artificial intelligence\", \"forest\", \"machine learning\"])\n",
    "## Are certain organizations more likely to mention ethics terms in conjunction with AI/ML? (Crawl)\n",
    "avg, freq = analyze_orgs(words.loc[(words[\"algorithm\"]>0) | (words[\"artificial intelligence\"]>0) | (words[\" ai \"]>0) | (words[\"machine learning\"]>0)], orgs = [\"raresites\", \"davidsuzuki\", \"grandriver\", \"wwf\", \"ecotrust\", \"natureconservancy\"], terms = [\"privacy\", \"surveillance\"])\n",
    "avg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "23e49ba2-c66f-43e5-bd17-4ae472281e78",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ai  algorithm\n",
      "Zeros: 23084, 93%\n",
      "algorithm but not  ai : 321, 1%\n",
      " ai  but not algorithm: 833, 3%\n",
      "Together: 572, 2%\n",
      "\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ai</th>\n",
       "      <th>algorithm</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ai</th>\n",
       "      <td>1.00000</td>\n",
       "      <td>0.12221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>algorithm</th>\n",
       "      <td>0.12221</td>\n",
       "      <td>1.00000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               ai   algorithm\n",
       " ai        1.00000    0.12221\n",
       "algorithm  0.12221    1.00000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel=' ai ', ylabel='algorithm'>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "algorithm  ai \n",
      "Zeros: 23084, 93%\n",
      " ai  but not algorithm: 833, 3%\n",
      "algorithm but not  ai : 321, 1%\n",
      "Together: 572, 2%\n",
      "\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>algorithm</th>\n",
       "      <th>ai</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>algorithm</th>\n",
       "      <td>1.00000</td>\n",
       "      <td>0.12221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ai</th>\n",
       "      <td>0.12221</td>\n",
       "      <td>1.00000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           algorithm      ai \n",
       "algorithm    1.00000  0.12221\n",
       " ai          0.12221  1.00000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='algorithm', ylabel=' ai '>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Are certain topics associated with other (e.g. forest and AI vs mountain and AI?) (Twitter, Google, and Crawl)\n",
    "from web_observatory import analyze_term_correlations\n",
    "#analyze_term_correlations(words, [\"artificial intelligence\", \"river\", \" ai \"])\n",
    "analyze_term_correlations(words, [\" ai \", \"algorithm\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "91b4dbe0-5dd7-4089-9422-a049046ee26e",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}