OA in the news.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import requests\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import random\n",
    "import json\n",
    "import re\n",
    "\n",
    "import datetime\n",
    "\n",
    "fetch_oadoi = False\n",
    "fetch_crossref = False\n",
    "fetch_altmetric = False\n",
    "\n",
    "%matplotlib inline\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('data/cancer_news_stories.txt', sep='\\t', header=None)\n",
    "df.columns = ['created_at', 'moreover_url', 'url', 'title', 'snippet', 'altmetric_ids', 'dois', 'pmids']\n",
    "for ids in ['altmetric_ids', 'dois', 'pmids']:\n",
    "    df[ids] = df[ids].map(lambda x: ','.join(set(str(x).split(','))) if x == x else np.nan)\n",
    "df.index.name = 'story_id'\n",
    "df['created_at'] = pd.to_datetime(df.created_at)\n",
    "\n",
    "df = df[~df.dois.isnull()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df = df[df.created_at >= '2016-01-01']\n",
    "df = df[df.created_at < '2017-01-01']\n",
    "# df = df[df.has_cancer]\n",
    "# del df['has_cancer']\n",
    "\n",
    "df2 = pd.DataFrame(df.dois.str.split(',').tolist(), index=df.index).stack()\n",
    "df2.name = 'doi'\n",
    "df2.index = df2.index.droplevel(1)\n",
    "df = df.join(df2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def get_oaDOI(doi):\n",
    "    global counter\n",
    "    url = 'https://api.oadoi.org/%s' % doi\n",
    "    headers = {'Content-Type': 'application/json', \n",
    "       'Accept': 'application/json'}\n",
    "    \n",
    "    counter = counter + 1\n",
    "    if counter % 10 == 0:\n",
    "        print(counter)\n",
    "    response = None\n",
    "    try: \n",
    "        response = requests.get(url, headers=headers)\n",
    "        if response.status_code == 200:\n",
    "            return response.text\n",
    "    except Exception as e:\n",
    "        print(e)\n",
    "        print(response)\n",
    "        pass\n",
    "    return False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Stefanie \n",
    "# df2 = pd.read_csv('data/DOI_OAstatus.txt', sep='\\t')\n",
    "# df2.set_index('DOI', inplace=True)\n",
    "# df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "counter = 0\n",
    "\n",
    "def get_altmetric(altmetric_id):\n",
    "    global counter\n",
    "    url = 'http://api.altmetric.com/v1/doi/%s' % altmetric_id\n",
    "    headers = {'content-type': 'application/json'}\n",
    "    response = requests.post(url, headers=headers)\n",
    "    counter = counter + 1\n",
    "    if counter % 50 == 0:\n",
    "        print (counter)\n",
    "    try: \n",
    "        response = requests.get(url, headers=headers)\n",
    "        if response.status_code == 200:\n",
    "            return response.json()\n",
    "        else:\n",
    "            return False\n",
    "    except: \n",
    "        return False    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "counter = 0\n",
    "\n",
    "def get_crossref(doi):\n",
    "    global counter\n",
    "    url = 'https://api.crossref.org/works/%s' % doi\n",
    "    headers = {'content-type': 'application/json'}\n",
    "    response = requests.post(url, headers=headers)\n",
    "    counter = counter + 1\n",
    "    if counter % 50 == 0:\n",
    "        print (counter)\n",
    "    try: \n",
    "        response = requests.get(url, headers=headers)\n",
    "        if response.status_code == 200:\n",
    "            return response.json()\n",
    "        else:\n",
    "            return False\n",
    "    except: \n",
    "        return False    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "since:  2016-01-01 00:00:00\n",
      "until:  2016-12-31 19:03:32\n"
     ]
    }
   ],
   "source": [
    "# Dates in dataset\n",
    "print(\"since: \", df.created_at.min())\n",
    "print(\"until: \", df.created_at.max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "stories:  67236\n",
      "articles:  11523\n"
     ]
    }
   ],
   "source": [
    "# Number of stories and articles\n",
    "print(\"stories: \", len(df))\n",
    "print(\"articles: \", len(df.doi.unique()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    11523.000000\n",
       "mean         5.834939\n",
       "std         14.163605\n",
       "min          1.000000\n",
       "25%          1.000000\n",
       "50%          1.000000\n",
       "75%          5.000000\n",
       "max        270.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# average number of stories\n",
    "df.groupby('doi').size().describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# oaDOIresponses = filter(lambda x: type(x) == dict, dois_df)\n",
    "# print \"Discarding %s DOIs because could not fech. That's %.2f of all DOIs\" % (len(dois_df) - len(oaDOIresponses), ((len(dois_df) - len(oaDOIresponses))*100.0)/len(dois_df))\n",
    "# oaDOI = pd.DataFrame.from_records(oaDOIresponses, index='doi')\n",
    "# # print \"Discarding %s DOIs because of errors. That's %.2f of all DOIs\" % (len(oaDOI[~oaDOI.error.isnull()]), len(oaDOI[~oaDOI.error.isnull()]) * 100.0 / len(oaDOI))\n",
    "# # oaDOI = oaDOI[oaDOI.error.isnull()]\n",
    "# oaDOI.oa_color = oaDOI.oa_color.fillna('black')\n",
    "# del oaDOI['url']\n",
    "\n",
    "# df = df.merge(oaDOI, left_on='doi', right_index=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "if fetch_oadoi: \n",
    "    oaDOI = df['doi'].to_frame()\n",
    "    oaDOI['oaDOI_response'] = oaDOI.doi.map(get_oaDOI)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>doi</th>\n",
       "      <th>oa_color</th>\n",
       "      <th>free_fulltext_url</th>\n",
       "      <th>evidence</th>\n",
       "      <th>content</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10.1001/archderm.148.1.122</td>\n",
       "      <td>black</td>\n",
       "      <td>NaN</td>\n",
       "      <td>closed</td>\n",
       "      <td>{\"doi\": \"10.1001/archderm.148.1.122\", \"url\": \"...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.1001/archdermatol.2009.308</td>\n",
       "      <td>black</td>\n",
       "      <td>NaN</td>\n",
       "      <td>closed</td>\n",
       "      <td>{\"doi\": \"10.1001/archdermatol.2009.308\", \"url\"...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10.1001/archdermatol.2011.173</td>\n",
       "      <td>black</td>\n",
       "      <td>NaN</td>\n",
       "      <td>closed</td>\n",
       "      <td>{\"doi\": \"10.1001/archdermatol.2011.173\", \"url\"...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10.1001/archgenpsychiatry.2010.116</td>\n",
       "      <td>black</td>\n",
       "      <td>NaN</td>\n",
       "      <td>closed</td>\n",
       "      <td>{\"doi\": \"10.1001/archgenpsychiatry.2010.116\", ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10.1001/archinte.161.2.277</td>\n",
       "      <td>black</td>\n",
       "      <td>NaN</td>\n",
       "      <td>closed</td>\n",
       "      <td>{\"doi\": \"10.1001/archinte.161.2.277\", \"url\": \"...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  doi oa_color free_fulltext_url evidence  \\\n",
       "0          10.1001/archderm.148.1.122    black               NaN   closed   \n",
       "1       10.1001/archdermatol.2009.308    black               NaN   closed   \n",
       "2       10.1001/archdermatol.2011.173    black               NaN   closed   \n",
       "3  10.1001/archgenpsychiatry.2010.116    black               NaN   closed   \n",
       "4          10.1001/archinte.161.2.277    black               NaN   closed   \n",
       "\n",
       "                                             content  \n",
       "0  {\"doi\": \"10.1001/archderm.148.1.122\", \"url\": \"...  \n",
       "1  {\"doi\": \"10.1001/archdermatol.2009.308\", \"url\"...  \n",
       "2  {\"doi\": \"10.1001/archdermatol.2011.173\", \"url\"...  \n",
       "3  {\"doi\": \"10.1001/archgenpsychiatry.2010.116\", ...  \n",
       "4  {\"doi\": \"10.1001/archinte.161.2.277\", \"url\": \"...  "
      ]
     },
     "execution_count": 206,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# OLD\n",
    "# oaDOI = pd.read_csv('data/juan_rerun.csv')\n",
    "# oaDOI.oa_color = oaDOI.oa_color.fillna('black')\n",
    "# oaDOI.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# NEW\n",
    "oaDOI = pd.read_csv('data/oaDOI_20170615.csv')\n",
    "del oaDOI['story_id']\n",
    "oaDOI['json'] = oaDOI.oaDOI_response.map(lambda x: json.loads(x) if len(x) > 5 else False)\n",
    "\n",
    "def get_oadoi_field(x, field):\n",
    "    try: \n",
    "        return x['results'][0][field]\n",
    "    except:\n",
    "        return None\n",
    "oaDOI['oa_color'] = oaDOI.json.map(lambda x: get_oadoi_field(x, 'oa_color'))\n",
    "oaDOI['oa_color'].fillna('black', inplace=True)\n",
    "oaDOI['free_fulltext_url'] = oaDOI.json.map(lambda x: get_oadoi_field(x, 'free_fulltext_url'))\n",
    "oaDOI['evidence'] = oaDOI.json.map(lambda x: get_oadoi_field(x, 'evidence'))\n",
    "\n",
    "del oaDOI['oaDOI_response']\n",
    "del oaDOI['json']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "oa_color\n",
       "black    6997\n",
       "blue      286\n",
       "gold     1314\n",
       "green    2926\n",
       "dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "oaDOI.groupby('oa_color').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "oaDOI[oaDOI.oa_color == 'black'].doi.to_csv('data/cancer_dois_in_news_closed.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "oa_color\n",
       "black    60.722034\n",
       "blue      2.481993\n",
       "gold     11.403280\n",
       "green    25.392693\n",
       "dtype: float64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "oaDOI.groupby('oa_color').size().divide(.01*len(oaDOI))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df = df.merge(oaDOI, left_on='doi', right_on='doi')\n",
    "df.set_index('doi', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df.to_csv('data/for_lauren.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# WARNING: fetches from CrossRef and Altmetric\n",
    "# oaDOI['crossref'] = oaDOI.index.map(get_crossref)\n",
    "# oaDOI['crossref'] = oaDOI.crossref.map(lambda r: r['message'] if r else None)\n",
    "# oaDOI['crossref_json'] = oaDOI.crossref.map(json.dumps)\n",
    "# oaDOI['altmetric'] = oaDOI.index.map(get_altmetric)\n",
    "# oaDOI['altmetric-published-date'] = oaDOI.altmetric.map(lambda x: x['published_on'] if x and 'published_on' in x.keys() else None)\n",
    "# oaDOI['altmetric-published-date'] = pd.to_datetime(oaDOI['altmetric-published-date'].multiply(1e9))\n",
    "# oaDOI['altmetric_json'] = oaDOI.altmetric.map(json.dumps)\n",
    "# oaDOI.to_csv('data/oaDOI_withcrossref_and_altmetric.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 212,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# oaDOI.to_csv('data/oaDOI_withcrossref_and_altmetric.csv', encoding='utf8')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "crossref_altmetric = pd.read_csv('data/oaDOI_withcrossref_and_altmetric.csv')\n",
    "crossref_altmetric = crossref_altmetric.set_index('doi')\n",
    "crossref_altmetric = crossref_altmetric[['crossref_json', 'altmetric-published-date']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df = df.merge(crossref_altmetric, left_index=True, right_index=True)\n",
    "df.to_csv('data/oaDOI_withcrossref_and_altmetric_rerun_20170615.csv', encoding='utf8')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "del df['crossref_json']\n",
    "del df['dois']\n",
    "del df['pmids']\n",
    "del df['altmetric_ids']\n",
    "df.to_excel('data/oaDOI_withcrossref_and_altmetric_rerun_20170617_small.xlsx')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "oaDOI = df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def get_date(c):\n",
    "    c = json.loads(c)\n",
    "    if not c:\n",
    "        return False\n",
    "    try: \n",
    "        if 'issued' in c:\n",
    "            d = c['issued']\n",
    "    except:\n",
    "        print(type(c))\n",
    "        raise\n",
    "#     if 'published-online' in c:\n",
    "#         d = c['published-online']\n",
    "#     elif  'published-print' in c: \n",
    "#         d = c['published-print']\n",
    "#     else:\n",
    "#         d = False # c['created']\n",
    "    \n",
    "    try: \n",
    "        d = d['date-parts'][0]\n",
    "        if len(d) == 1:\n",
    "            return False\n",
    "        if len(d) == 2:\n",
    "            return datetime.datetime(d[0], d[1], 1)\n",
    "    \n",
    "        return datetime.datetime(d[0], d[1], d[2])\n",
    "    except:\n",
    "        return False\n",
    "    \n",
    "\n",
    "oaDOI['published-date'] = oaDOI.crossref_json.map(get_date)\n",
    "len(oaDOI[oaDOI['published-date'].notnull()])*100./len(oaDOI)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def get_min_date(x):\n",
    "    if x['published-date'] and x['altmetric-published-date']:\n",
    "        return min(x['published-date'], x['altmetric-published-date'])\n",
    "    \n",
    "    if x['published-date']:\n",
    "        return x['published-date']\n",
    "    \n",
    "    if x['altmetric-published-date']:\n",
    "        return x['altmetric-published-date']\n",
    "\n",
    "# change False to None\n",
    "oaDOI['published-date'] = oaDOI['published-date'].map(lambda x: x if x and x != 'False' else None)\n",
    "oaDOI['altmetric-published-date'] = oaDOI['altmetric-published-date'].map(lambda x: x if x and x != 'False' else None)\n",
    "\n",
    "oaDOI['published-date'] = pd.to_datetime(oaDOI['published-date'])\n",
    "oaDOI['altmetric-published-date'] = pd.to_datetime(oaDOI['altmetric-published-date'])\n",
    "\n",
    "oaDOI['earliest-pub-date'] = oaDOI.apply(get_min_date, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 219,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dropping 1793 with no dates (2.7%)\n"
     ]
    }
   ],
   "source": [
    "# try: \n",
    "#     del df['earliest-pub-date']\n",
    "# except:\n",
    "#     pass\n",
    "# df = df.merge(oaDOI[['earliest-pub-date']], left_index=True, right_index=True)\n",
    "df['days_old'] = df.apply(lambda row: row['created_at'] - row['earliest-pub-date'] if row['earliest-pub-date'] else None, axis=1)\n",
    "print(\"Dropping %s with no dates (%.1f%%)\" % (len(df[df.days_old.isnull()]), len(df[df.days_old.isnull()])*100/len(df)))\n",
    "df = df[df.days_old.notnull()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count                        65093\n",
       "mean      340 days 13:18:50.108905\n",
       "std      1156 days 08:06:15.190200\n",
       "min             -29 days +04:00:00\n",
       "25%                1 days 15:49:02\n",
       "50%               14 days 20:05:16\n",
       "75%               96 days 11:00:25\n",
       "max            42878 days 20:37:08\n",
       "Name: days_old, dtype: object"
      ]
     },
     "execution_count": 220,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[(df.days_old.map(lambda x: x.days) >= -30)].days_old.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 221,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count                      55325\n",
       "mean     37 days 01:17:39.871938\n",
       "std      68 days 01:01:11.798595\n",
       "min           -29 days +04:00:00\n",
       "25%              1 days 11:06:00\n",
       "50%              7 days 21:21:56\n",
       "75%             36 days 08:13:00\n",
       "max            365 days 20:37:28\n",
       "Name: days_old, dtype: object"
      ]
     },
     "execution_count": 221,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[(df.days_old.map(lambda x: x.days) >= -30)  & (df.days_old.map(lambda x: x.days) <= 365)].days_old.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tableau20_rgb = [(31, 119, 180), (174, 199, 232), (255, 127, 14), (255, 187, 120),  \n",
    "             (44, 160, 44), (152, 223, 138), (214, 39, 40), (255, 152, 150),  \n",
    "             (148, 103, 189), (197, 176, 213), (140, 86, 75), (196, 156, 148),  \n",
    "             (227, 119, 194), (247, 182, 210), (127, 127, 127), (199, 199, 199),  \n",
    "             (188, 189, 34), (219, 219, 141), (23, 190, 207), (158, 218, 229)] \n",
    "\n",
    "# Scale the RGB values to the [0, 1] range, which is the format matplotlib accepts.  \n",
    "tableau20 = {}\n",
    "for i, (r,g,b) in enumerate(tableau20_rgb):  \n",
    "    tableau20[i] = (r / 255., g / 255., b / 255.)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1UAAAH/CAYAAAC/5/dnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQRAEQRAEY8H/A+DanJK7II8GAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115f4c518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import locale\n",
    "locale.setlocale(locale.LC_ALL, 'en_US')\n",
    "\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (12.0, 8.0)\n",
    "fig, ax = plt.subplots()\n",
    "df.days_old.map(lambda x: x.days).hist(bins=range(-30, 390, 30), color=tableau20[0], linewidth=0, ax=ax)\n",
    "ax.set_xlim([-30, 360])\n",
    "ax.set_ylim([0, 32000])\n",
    "ax.xaxis.grid(False)\n",
    "\n",
    "ax.yaxis.grid(True)\n",
    "ax.set_axisbelow(True)\n",
    "\n",
    "# Remove the plot frame lines. They are unnecessary chartjunk.  \n",
    "ax.spines[\"top\"].set_visible(False)  \n",
    "ax.spines[\"bottom\"].set_color('grey')\n",
    "ax.spines[\"right\"].set_visible(False)  \n",
    "ax.spines[\"left\"].set_color('grey')  \n",
    "\n",
    "ax.tick_params(axis=\"both\", which=\"both\", bottom=\"off\", top=\"off\",  \n",
    "                labelbottom=\"on\", left=\"off\", right=\"off\", labelleft=\"on\") \n",
    "\n",
    "plt.yticks(range(0,32500, 5000), [str(x) for x in range(0,32, 5)], color='grey', fontsize=14)  \n",
    "ax.set_ylabel('Number of stories\\n(Thousands)', color='grey', fontsize=16)\n",
    "\n",
    "ticks = plt.xticks([-15] + list(range(15, 365, 30)), ['One month before'] + ['First', 'Second', 'Third', 'Fourth', 'Fifth', 'Sixth', 'Seventh', 'Eighth', 'Ninth', 'Tenth', 'Eleventh', 'Twelfth'], rotation=45, color='grey', ha='right', fontsize=14)\n",
    "ax.set_xlabel('Months since publication', color='grey', fontsize=16)\n",
    "\n",
    "ax.set_title('How long until a news story is written?', fontsize=20)\n",
    "plt.tight_layout(rect=[0, .1, 1, 1])\n",
    "\n",
    "plt.savefig('data/months_since_publication_v2.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/site-packages/ipykernel/__main__.py:1: 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: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  if __name__ == '__main__':\n"
     ]
    }
   ],
   "source": [
    "df['bins'] = pd.cut(df.days_old.map(lambda x: x.days), range(-30, 365, 30))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bins\n",
       "(-30, 0]      10713\n",
       "(0, 30]       29329\n",
       "(30, 60]       5821\n",
       "(60, 90]       2590\n",
       "(90, 120]      1789\n",
       "(120, 150]      993\n",
       "(150, 180]      777\n",
       "(180, 210]      687\n",
       "(210, 240]      601\n",
       "(240, 270]      566\n",
       "(270, 300]      427\n",
       "(300, 330]      565\n",
       "(330, 360]      390\n",
       "dtype: int64"
      ]
     },
     "execution_count": 192,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('bins').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'DataFrame' object has no attribute 'crossref'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-193-dce52622139b>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0moaDOI\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'funder'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moaDOI\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcrossref\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'funder'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mr\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0;34m'funder'\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mr\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mfunders\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moaDOI\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m~\u001b[0m\u001b[0moaDOI\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfunder\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0misnull\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfunder\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'name'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;34m'name'\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mf\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mf\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mfunders\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mfunders\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m'National Institute'\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/pandas/core/generic.py\u001b[0m in \u001b[0;36m__getattr__\u001b[0;34m(self, name)\u001b[0m\n\u001b[1;32m   2742\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mname\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_info_axis\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2743\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2744\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mobject\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__getattribute__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2745\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2746\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__setattr__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mAttributeError\u001b[0m: 'DataFrame' object has no attribute 'crossref'"
     ]
    }
   ],
   "source": [
    "oaDOI['funder'] = oaDOI.crossref.map(lambda r: r['funder'] if r and 'funder' in r else None)\n",
    "\n",
    "funders = oaDOI[~oaDOI.funder.isnull()].funder.map(lambda x: [f['name'] if 'name' in f else None for f in x])\n",
    "funders[funders.map(lambda f: 'National Institute' in f)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df = df.reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# number of stories by different limits\n",
    "df_lims = {}\n",
    "for lim in [0, 10, 25, 50]:\n",
    "    df_lims[lim] = df.groupby('doi').filter(lambda x: len(x) >= lim)\n",
    "    print \"At least %s: %s\" % (lim, len(df_lims[lim].doi.unique()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df = df.reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df = df.merge(oaDOI, left_on='doi', right_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "oa_mentions = {}\n",
    "for lim in range(0, 60, 10):\n",
    "    df2 = df.groupby('doi').filter(lambda x: len(x) >= lim)\n",
    "    oa_color = df2[['doi', 'oa_color']].drop_duplicates().set_index('doi')\n",
    "    oa_color = oa_color.groupby('oa_color').size().multiply(100.0).divide(len(oa_color))\n",
    "    oa_mentions[lim] = oa_color['gold']\n",
    "    print \"At least %s: %s articles of which %.2f%% are OA\" % (lim, len(df2.doi.unique()), oa_mentions[lim])\n",
    "#     print oa_color.to_string()\n",
    "#     print"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df[['doi', 'oa_color']].drop_duplicates().set_index('doi').groupby('oa_color').size().divide(.01*11044).map(lambda x: \"%.2f%%\" % x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "oa_color = df[['doi', 'oa_color']].drop_duplicates().set_index('doi')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df[['doi', 'oa_color']].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "print len(df[['doi', 'oa_color']])\n",
    "print len(df[['doi', 'oa_color']].drop_duplicates())\n",
    "print len(oa_color)\n",
    "print oa_color.groupby('oa_color').size().divide(len(oa_color)*.01)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df[['doi', 'oa_color', 'evidence']].drop_duplicates().groupby(['oa_color', 'evidence']).size().sort_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "num_mentions = df.groupby('doi').size()\n",
    "num_mentions.name = 'num_mentions'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "oa_color = df.set_index('doi').join(num_mentions).reset_index()[['doi', 'oa_color', 'num_mentions']].drop_duplicates().set_index('doi')\n",
    "oa_color = oa_color[~oa_color.num_mentions.isnull()]\n",
    "oa_color['doi_as_link'] = oa_color.index.map(lambda doi: \"http://doi.org/%s\" % doi)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "oa_color.sort_values('num_mentions', ascending=False).to_csv('data/news_story_dois_oastatus.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "gb = oa_color.groupby(['num_mentions', 'oa_color'])\n",
    "# gb.size().unstack().apply(lambda row: row['gold']*100.0/sum(row), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "pylab.rcParams['figure.figsize'] = (12.0, 8.0)\n",
    "df3 = gb.size().unstack().fillna(0).apply(lambda row: row['gold']*100.0/sum(row), axis=1)\n",
    "area = np.pi * (10+log(oa_color.groupby('num_mentions').size().divide(len(oa_color))))**2\n",
    "fig, ax = plt.subplots()\n",
    "plt.scatter(df3.index, df3, s=area)\n",
    "ax.set_xlim([0,300])\n",
    "ax.set_ylim([-.5, 105])\n",
    "ax.set_xlabel('exactly x mentions')\n",
    "ax.set_ylabel('% OA')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "df3 = gb.size().unstack().fillna(0).sort_index(ascending=False).cumsum().sort_index()\n",
    "\n",
    "# ax = df3.sum(axis=1).plot.bar()\n",
    "area = np.pi*log(df3.sum(axis=1))**2\n",
    "\n",
    "# errors = df3.apply(lambda row: row['gold']/float(sum(row)), axis=1).map(lambda x: x*(1-x)).divide(df3.sum(axis=1)).map(np.sqrt).multiply(100)\n",
    "df3.apply(lambda row: row['gold']*100.0/sum(row), axis=1).reset_index().plot.scatter(x='num_mentions', y=0, title=\"% OA by minimum number of news mentions\", s=30, xlim=[0,150], alpha=.5, ax=ax)\n",
    "\n",
    "ax.set_xlim([0, 165])\n",
    "# ax.set_ylim([0, 32000])\n",
    "ax.xaxis.grid(False)\n",
    "\n",
    "ax.yaxis.grid(True)\n",
    "ax.set_axisbelow(True)\n",
    "\n",
    "# Remove the plot frame lines. They are unnecessary chartjunk.  \n",
    "ax.spines[\"top\"].set_visible(False)  \n",
    "ax.spines[\"bottom\"].set_color('grey')\n",
    "ax.spines[\"right\"].set_visible(False)  \n",
    "ax.spines[\"left\"].set_color('grey')  \n",
    "\n",
    "ax.tick_params(axis=\"both\", which=\"both\", bottom=\"off\", top=\"off\",  \n",
    "                labelbottom=\"on\", left=\"off\", right=\"off\", labelleft=\"on\") \n",
    "\n",
    "\n",
    "\n",
    "# ax.set_xlim([0,df3.index.max()+5])\n",
    "\n",
    "# ax2 = ax.twinx()\n",
    "# # area = np.pi * (15*())**2\n",
    "\n",
    "ax.set_ylim([0,105])\n",
    "ax.set_xlabel('at least x mentions')\n",
    "ax.set_ylabel('% OA')\n",
    "\n",
    "plt.savefig('data/oa_by_min_mentions.png')\n",
    "\n",
    "# # ax2 = plt.scatter(df3.index, , s=30)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Number of Gold OA for articles with at least 50 mentions\n",
    "df3 = oa_color[oa_color.num_mentions >= 50].groupby(['num_mentions', 'oa_color']).size().unstack().fillna(0).sort_index(ascending=False).cumsum().sort_index()\n",
    "print \"%.2f%% gold OA\" % (df3['gold'].sum() * 100.0 / df3.sum(axis=1).sum())\n",
    "print \"%.2f%% green\" % (df3['green'].sum() * 100.0 / df3.sum(axis=1).sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from urlparse import urlparse\n",
    "df['url'] = df.apply(lambda row: row['url'] if row['url'] == row['url'] else row['moreover_url'], axis=1)\n",
    "df['domain'] = df.url.map(lambda url: re.sub('^www\\.', '', urlparse(url).netloc) if url == url else np.nan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df.groupby('domain').size().describe()['75%']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Top 25% of domain names\n",
    "df.groupby('domain').filter(lambda x: len(x) > 100).groupby('domain').size().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "domains = ['nytimes.com', 'washingtonpost.com', 'slate.com', 'huffingtonpost.com', 'theatlantic.com', 'theguardian.com', 'salon.com', 'latimes.com', 'foxnews.com', 'npr.org']\n",
    "\n",
    "for d in domains: \n",
    "    domain_df = df[df.domain.str.contains(d)]\n",
    "    print \"Domain: \", d, [x for x in domain_df.domain.unique()]\n",
    "    print \"%s articles\" % len(domain_df)\n",
    "    print \"%.2f%% OA\" % domain_df.groupby('oa_color').size().multiply(100.0).divide(len(domain_df))['gold']\n",
    "    print\n",
    "del domain_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "sample_10p.groupby('classification').size().multiply(100.0).div(len(sample_10p)).plot.bar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "sample_10p.groupby('dissmin_classification').size().multiply(100.0).div(len(sample_10p))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "sample_10p.groupby('classification').size().multiply(100.0).div(len(sample_10p))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "first_story = df.sort_values(by='posted_on:conv').groupby('citation:doi').first()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df3 = oa_color[oa_color.num_mentions >= 100].groupby(['num_mentions', 'oa_color']).size().unstack().fillna(0).sort_index(ascending=False).cumsum().sort_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df3.apply(lambda row: row['gold']/float(sum(row)), axis=1).map(lambda x: x*(1-x)).divide(df3.sum(axis=1)).map(np.sqrt).multiply(100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "oa_color = df.set_index('doi').join(num_mentions).reset_index()[['doi', 'oa_color', 'num_mentions']].drop_duplicates().set_index('doi')\n",
    "oa_color = oa_color[~oa_color.num_mentions.isnull()]\n",
    "\n",
    "oa_color.to_csv('data/top50news_oadoi_status.csv', sep=\"\\t\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "\n",
    "b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "gb = first_story.join(sample_10p.set_index('doi'))\n",
    "gb['year'] = gb['posted_on:conv'].map(lambda x: x.year)\n",
    "gb = gb[gb.year > 2011]\n",
    "gb = gb[gb.classification.notnull()].groupby(['year', 'classification']).size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tmp = gb.multiply(100.0).divide(gb.groupby(level=0).sum()).unstack()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tmp.index = [2012,2013,2014]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "response = get_crossref('http://api.crossref.org/works/10.1016/j.neuroscience.2015.10.035')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# for stef\n",
    "len(df.index.tolist())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "responses = []\n",
    "for i in range(0, 40, 25):    \n",
    "    dois = df.iloc[i:i+25].apply(lambda x: ''.join(x.index)).tolist()\n",
    "    responses += get_oaDOIs(dois)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "len(responses)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "oaDOI = pd.DataFrame.from_records(responses, index='doi')\n",
    "print \"Discarding %s DOIs because of errors. That's %.2f of all DOIs\" % (len(oaDOI[~oaDOI.error.isnull()]), len(oaDOI[~oaDOI.error.isnull()]) * 100.0 / len(oaDOI))\n",
    "oaDOI = oaDOI[oaDOI.error.isnull()]\n",
    "oaDOI.oa_color = oaDOI.oa_color.fillna('black')\n",
    "\n",
    "df2 = df.join(oaDOI)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('data/oaDOI_withcrossref_and_altmetric.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df_new = df[[u'doi', u'doi_resolver', u'license', u'free_fulltext_url', u'is_free_to_read', u'is_boai_license', u'is_subscription_journal', u'oa_color', u'evidence', u'_title']].drop_duplicates().set_index('doi')\n",
    "print len(df_new)\n",
    "df_new.to_csv('data/new_oaDOI_responses.csv', encoding='utf8')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df_old = pd.read_csv('data/oaDOI_withcrossref.csv', encoding='utf8')\n",
    "df_old.set_index('doi', inplace=True)\n",
    "df_old.groupby('oa_color').size().divide(len(df_old)*.01)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df_new.merge(df_old, how='left', left_index=True, right_index=True, suffixes=('_new', '_old')).to_csv('data/oaDOIcomparison.csv', encoding='utf8')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "cp = df_new.merge(df_old, how='left', left_index=True, right_index=True, suffixes=('_new', '_old'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "cp[cp['oa_color_old'] != cp['oa_color_new']][['oa_color_new', 'evidence_new', 'oa_color_old', 'evidence_old']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df_compare = pd.read_csv('data/oaDOIcomparison.csv', encoding='utf8')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df_compare.groupby('oa_color_old').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df_change = df_compare[df_compare['oa_color_old'] == 'gold'][['evidence_old', 'oa_color_new', 'evidence_new']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df_change.groupby('oa_color_new').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "df_change[df_change.oa_color_new != 'gold'].groupby('evidence_new').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "df.merge(oaDOI, left_on='doi', right_on='doi').to_excel('data/for_lauren.xlsx')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}