{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import janitor\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from matplotlib.ticker import MaxNLocator\n",
    "import math\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "outputs": [],
   "source": [
    "outdir=\"WESTERN_CH_scope\"\n",
    "\n",
    "appln = pd.read_csv(f\"{outdir}/tls_201_scope.csv\")\n",
    "\n",
    "appln_title = pd.read_csv(f\"{outdir}/tls_202_scope.csv\")\n",
    "\n",
    "pers = pd.read_csv(f\"{outdir}/tls_206_scope.csv\")\n",
    "pers['psn_sector'] = pers['psn_sector'].fillna(\"UNKNOWN\")\n",
    "\n",
    "appln_pers = pd.read_csv(f\"{outdir}/tls_207_scope.csv\")\n",
    "\n",
    "appln_cpc = pd.read_csv(f\"{outdir}/tls_224_scope.csv\")"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "outputs": [
    {
     "data": {
      "text/plain": "203873"
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(appln)"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "outputs": [
    {
     "data": {
      "text/plain": "         appln_id appln_title_lg   \n106316  498640253             en  \\\n119852  511974583             en   \n193586  577006640             en   \n172207  556318748             en   \n117620  509549284             en   \n...           ...            ...   \n58791   448189845             en   \n119362  511604550             en   \n73722   471815906             en   \n45133   438311946             en   \n25978   414431520             en   \n\n                                              appln_title  \n106316                              DRAIN CLEANING DEVICE  \n119852        Antenna panel switching and beam indication  \n193586                                           Loft bed  \n172207                          Winch for Securing a Load  \n117620  TEMPERATURE CONTROL APPARATUS FORELECTRIC VEHI...  \n...                                                   ...  \n58791   Collaborative spectrum sensing in cognitive ra...  \n119362              CLIP-ON GLASSES WITH REPLACEABLE LENS  \n73722   Sensitized, photo-sensitive glass and its prod...  \n45133   PREPARATION OF 3,4-DIHYDRO-1,4-BENZOXAZEPIN-5(...  \n25978            - CRYSTAL OSCILLATOR WITH LOW-POWER MODE  \n\n[100 rows x 3 columns]",
      "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>appln_id</th>\n      <th>appln_title_lg</th>\n      <th>appln_title</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>106316</th>\n      <td>498640253</td>\n      <td>en</td>\n      <td>DRAIN CLEANING DEVICE</td>\n    </tr>\n    <tr>\n      <th>119852</th>\n      <td>511974583</td>\n      <td>en</td>\n      <td>Antenna panel switching and beam indication</td>\n    </tr>\n    <tr>\n      <th>193586</th>\n      <td>577006640</td>\n      <td>en</td>\n      <td>Loft bed</td>\n    </tr>\n    <tr>\n      <th>172207</th>\n      <td>556318748</td>\n      <td>en</td>\n      <td>Winch for Securing a Load</td>\n    </tr>\n    <tr>\n      <th>117620</th>\n      <td>509549284</td>\n      <td>en</td>\n      <td>TEMPERATURE CONTROL APPARATUS FORELECTRIC VEHI...</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>58791</th>\n      <td>448189845</td>\n      <td>en</td>\n      <td>Collaborative spectrum sensing in cognitive ra...</td>\n    </tr>\n    <tr>\n      <th>119362</th>\n      <td>511604550</td>\n      <td>en</td>\n      <td>CLIP-ON GLASSES WITH REPLACEABLE LENS</td>\n    </tr>\n    <tr>\n      <th>73722</th>\n      <td>471815906</td>\n      <td>en</td>\n      <td>Sensitized, photo-sensitive glass and its prod...</td>\n    </tr>\n    <tr>\n      <th>45133</th>\n      <td>438311946</td>\n      <td>en</td>\n      <td>PREPARATION OF 3,4-DIHYDRO-1,4-BENZOXAZEPIN-5(...</td>\n    </tr>\n    <tr>\n      <th>25978</th>\n      <td>414431520</td>\n      <td>en</td>\n      <td>- CRYSTAL OSCILLATOR WITH LOW-POWER MODE</td>\n    </tr>\n  </tbody>\n</table>\n<p>100 rows × 3 columns</p>\n</div>"
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "appln_title.sample(100)"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "outputs": [
    {
     "data": {
      "text/plain": "    appln_id appln_auth  appln_nr appln_kind appln_filing_date   \n0  330225325         EP  11150195         A         2011-01-05  \\\n1  330225397         EP  11150231         A         2011-01-05   \n2  330322632         EP  11150485         A         2011-01-10   \n3  330326785         EP  11150605         A         2011-01-11   \n4  330350961         EP  11150683         A         2011-01-12   \n\n   appln_filing_year appln_nr_original ipr_type receiving_office   \n0               2011          11150195       PI                   \\\n1               2011          11150231       PI                    \n2               2011          11150485       PI                    \n3               2011          11150605       PI                    \n4               2011          11150683       PI                    \n\n   internat_appln_id  ... earliest_pat_publn_id granted docdb_family_id   \n0                  0  ...             335277427       Y        43754737  \\\n1                  0  ...             335277736       Y        43619902   \n2                  0  ...             364719889       Y        43991052   \n3                  0  ...             335277720       N        43023665   \n4                  0  ...             364923578       N        43881056   \n\n  inpadoc_family_id  docdb_family_size  nb_citing_docdb_fam nb_applicants   \n0         330225325                  4                   16             1  \\\n1         330225397                  6                   56             1   \n2         330322632                  2                    5             1   \n3         328518903                  6                    9             1   \n4         330350961                  7                   13             2   \n\n   nb_inventors  appln_title_lg   \n0             1              en  \\\n1             9              en   \n2             2              en   \n3             3              en   \n4             5              en   \n\n                                         appln_title  \n0                       Beverage preparation machine  \n1  Screwdriving tool having a driving tool with a...  \n2  Method and system for recommending contextual ...  \n3  Apparatus and method for continuous casting of...  \n4  A method and an apparatus for treating at leas...  \n\n[5 rows x 28 columns]",
      "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>appln_id</th>\n      <th>appln_auth</th>\n      <th>appln_nr</th>\n      <th>appln_kind</th>\n      <th>appln_filing_date</th>\n      <th>appln_filing_year</th>\n      <th>appln_nr_original</th>\n      <th>ipr_type</th>\n      <th>receiving_office</th>\n      <th>internat_appln_id</th>\n      <th>...</th>\n      <th>earliest_pat_publn_id</th>\n      <th>granted</th>\n      <th>docdb_family_id</th>\n      <th>inpadoc_family_id</th>\n      <th>docdb_family_size</th>\n      <th>nb_citing_docdb_fam</th>\n      <th>nb_applicants</th>\n      <th>nb_inventors</th>\n      <th>appln_title_lg</th>\n      <th>appln_title</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>330225325</td>\n      <td>EP</td>\n      <td>11150195</td>\n      <td>A</td>\n      <td>2011-01-05</td>\n      <td>2011</td>\n      <td>11150195</td>\n      <td>PI</td>\n      <td></td>\n      <td>0</td>\n      <td>...</td>\n      <td>335277427</td>\n      <td>Y</td>\n      <td>43754737</td>\n      <td>330225325</td>\n      <td>4</td>\n      <td>16</td>\n      <td>1</td>\n      <td>1</td>\n      <td>en</td>\n      <td>Beverage preparation machine</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>330225397</td>\n      <td>EP</td>\n      <td>11150231</td>\n      <td>A</td>\n      <td>2011-01-05</td>\n      <td>2011</td>\n      <td>11150231</td>\n      <td>PI</td>\n      <td></td>\n      <td>0</td>\n      <td>...</td>\n      <td>335277736</td>\n      <td>Y</td>\n      <td>43619902</td>\n      <td>330225397</td>\n      <td>6</td>\n      <td>56</td>\n      <td>1</td>\n      <td>9</td>\n      <td>en</td>\n      <td>Screwdriving tool having a driving tool with a...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>330322632</td>\n      <td>EP</td>\n      <td>11150485</td>\n      <td>A</td>\n      <td>2011-01-10</td>\n      <td>2011</td>\n      <td>11150485</td>\n      <td>PI</td>\n      <td></td>\n      <td>0</td>\n      <td>...</td>\n      <td>364719889</td>\n      <td>Y</td>\n      <td>43991052</td>\n      <td>330322632</td>\n      <td>2</td>\n      <td>5</td>\n      <td>1</td>\n      <td>2</td>\n      <td>en</td>\n      <td>Method and system for recommending contextual ...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>330326785</td>\n      <td>EP</td>\n      <td>11150605</td>\n      <td>A</td>\n      <td>2011-01-11</td>\n      <td>2011</td>\n      <td>11150605</td>\n      <td>PI</td>\n      <td></td>\n      <td>0</td>\n      <td>...</td>\n      <td>335277720</td>\n      <td>N</td>\n      <td>43023665</td>\n      <td>328518903</td>\n      <td>6</td>\n      <td>9</td>\n      <td>1</td>\n      <td>3</td>\n      <td>en</td>\n      <td>Apparatus and method for continuous casting of...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>330350961</td>\n      <td>EP</td>\n      <td>11150683</td>\n      <td>A</td>\n      <td>2011-01-12</td>\n      <td>2011</td>\n      <td>11150683</td>\n      <td>PI</td>\n      <td></td>\n      <td>0</td>\n      <td>...</td>\n      <td>364923578</td>\n      <td>N</td>\n      <td>43881056</td>\n      <td>330350961</td>\n      <td>7</td>\n      <td>13</td>\n      <td>2</td>\n      <td>5</td>\n      <td>en</td>\n      <td>A method and an apparatus for treating at leas...</td>\n    </tr>\n  </tbody>\n</table>\n<p>5 rows × 28 columns</p>\n</div>"
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "appln_data = appln.merge(appln_title, on=\"appln_id\")\n",
    "appln_data.head()"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "outputs": [
    {
     "data": {
      "text/plain": "array(['EP', 'WO', 'LU', 'FI', 'NO', 'FR', 'GB', 'KR', 'ES', 'US', 'CA',\n       'DO', 'EC', 'DE', 'UY', 'IL', 'SV', 'PL', 'TR', 'CO', 'CR', 'TW',\n       'MA', 'PE', 'SG', 'CU', 'BE', 'DK', 'AR', 'AP', 'HR', 'MX', 'BR',\n       'EA', 'RU', 'AU', 'MC', 'HU', 'PT', 'NL', 'HN', 'AT', 'RO', 'SM',\n       'CH', 'SI', 'IS', 'CZ', 'HK', 'MD', 'JP', 'CN', 'RS', 'GT', 'UA',\n       'CL', 'SK', 'LT', 'PH', 'MY', 'IN', 'VN', 'TN', 'CY', 'GE', 'ZA',\n       'SE', 'ME', 'JO', 'NI', 'SA'], dtype=object)"
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "appln_data[\"appln_auth\"].unique()"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "outputs": [
    {
     "data": {
      "text/plain": "        person_id   appln_id  applt_seq_nr  invt_seq_nr\n0               1  340314532             1            0\n1               1  413601768             1            0\n2              21  332015605             1            0\n3              21  333490084             1            0\n4              21  335903805             1            0\n...           ...        ...           ...          ...\n274039   85719932  545918634             0            2\n274040   85720336  569409547             0            4\n274041   85720376  555215896             0            2\n274042   85720469  569304088             0            5\n274043   85720500  569495993             0            5\n\n[274044 rows x 4 columns]",
      "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>person_id</th>\n      <th>appln_id</th>\n      <th>applt_seq_nr</th>\n      <th>invt_seq_nr</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>1</td>\n      <td>340314532</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1</td>\n      <td>413601768</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>21</td>\n      <td>332015605</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>21</td>\n      <td>333490084</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>21</td>\n      <td>335903805</td>\n      <td>1</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    </tr>\n    <tr>\n      <th>274039</th>\n      <td>85719932</td>\n      <td>545918634</td>\n      <td>0</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>274040</th>\n      <td>85720336</td>\n      <td>569409547</td>\n      <td>0</td>\n      <td>4</td>\n    </tr>\n    <tr>\n      <th>274041</th>\n      <td>85720376</td>\n      <td>555215896</td>\n      <td>0</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>274042</th>\n      <td>85720469</td>\n      <td>569304088</td>\n      <td>0</td>\n      <td>5</td>\n    </tr>\n    <tr>\n      <th>274043</th>\n      <td>85720500</td>\n      <td>569495993</td>\n      <td>0</td>\n      <td>5</td>\n    </tr>\n  </tbody>\n</table>\n<p>274044 rows × 4 columns</p>\n</div>"
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "appln_pers"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "outputs": [
    {
     "data": {
      "text/plain": "        person_id                person_name        person_name_orig_lg   \n0               1          Nokia Corporation          Nokia Corporation  \\\n1             128  Nokia Siemens Networks Oy  Nokia Siemens Networks Oy   \n2         5217785          Nokia Corporation          Nokia Corporation   \n3         5217811          Nokia Corporation          Nokia Corporation   \n4         5232170  Nokia Siemens Networks Oy  Nokia Siemens Networks Oy   \n...           ...                        ...                        ...   \n112235   85719932             VIKSTREM, Erik             ВИКСТРЁМ, Эрик   \n112236   85720336            HWANG, LING-CHI            HWANG, LING-CHI   \n112237   85720376                 LI, I Chan                 LI, I Chan   \n112238   85720469           TING, Chia Ching           TING, Chia Ching   \n112239   85720500             WANG, YU-CHEIH             WANG, YU-CHEIH   \n\n                      person_address person_ctry_code   nuts  nuts_level   \n0       Keilalahdentie 4,02150 Espoo               FI  FI1B1           3  \\\n1           Karaportti 3,02610 Espoo               FI  FI1B1           3   \n2                              Espoo               FI     FI           0   \n3                                NaN               FI     FI           0   \n4                              Espoo               FI     FI           0   \n...                              ...              ...    ...         ...   \n112235                           NaN               SE     SE           0   \n112236                           NaN               TW    NaN           9   \n112237                           NaN               TW    NaN           9   \n112238                            TW               TW    NaN           9   \n112239                           NaN               TW    NaN           9   \n\n        doc_std_name_id               doc_std_name     psn_id   \n0                     1                 NOKIA CORP   23782051  \\\n1                   112  NOKIA SIEMENS NETWORKS OY   23782129   \n2                     1                 NOKIA CORP   23782051   \n3                     1                 NOKIA CORP   23782051   \n4                   112  NOKIA SIEMENS NETWORKS OY   23782129   \n...                 ...                        ...        ...   \n112235         38919340              VIKSTREM ERIK  185719932   \n112236         35599384             HWANG LING-CHI  185720336   \n112237         38707281                  LI I CHAN  185720376   \n112238         23937900            TING CHIA CHING  185720469   \n112239         38204835              WANG YU-CHEIH  185720500   \n\n                 psn_name  psn_level psn_sector     han_id          han_name   \n0       NOKIA CORPORATION          2    COMPANY    2125445        NOKIA CORP  \\\n1          NOKIA NETWORKS          2    COMPANY    2125445        NOKIA CORP   \n2       NOKIA CORPORATION          2    COMPANY    2125445        NOKIA CORP   \n3       NOKIA CORPORATION          2    COMPANY    2125445        NOKIA CORP   \n4          NOKIA NETWORKS          2    COMPANY    2125445        NOKIA CORP   \n...                   ...        ...        ...        ...               ...   \n112235     VIKSTREM, Erik          0    UNKNOWN  185719932    VIKSTREM, Erik   \n112236    HWANG, LING-CHI          0    UNKNOWN  185720336   HWANG, LING-CHI   \n112237         LI, I Chan          0    UNKNOWN  185720376        LI, I Chan   \n112238   TING, Chia Ching          0    UNKNOWN  185720469  TING, Chia Ching   \n112239     WANG, YU-CHEIH          0    UNKNOWN  185720500    WANG, YU-CHEIH   \n\n        han_harmonized psn_sector_primary  \n0                    2            COMPANY  \n1                    2            COMPANY  \n2                    2            COMPANY  \n3                    2            COMPANY  \n4                    2            COMPANY  \n...                ...                ...  \n112235               0            UNKNOWN  \n112236               0            UNKNOWN  \n112237               0            UNKNOWN  \n112238               0            UNKNOWN  \n112239               0            UNKNOWN  \n\n[112240 rows x 17 columns]",
      "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>person_id</th>\n      <th>person_name</th>\n      <th>person_name_orig_lg</th>\n      <th>person_address</th>\n      <th>person_ctry_code</th>\n      <th>nuts</th>\n      <th>nuts_level</th>\n      <th>doc_std_name_id</th>\n      <th>doc_std_name</th>\n      <th>psn_id</th>\n      <th>psn_name</th>\n      <th>psn_level</th>\n      <th>psn_sector</th>\n      <th>han_id</th>\n      <th>han_name</th>\n      <th>han_harmonized</th>\n      <th>psn_sector_primary</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>1</td>\n      <td>Nokia Corporation</td>\n      <td>Nokia Corporation</td>\n      <td>Keilalahdentie 4,02150 Espoo</td>\n      <td>FI</td>\n      <td>FI1B1</td>\n      <td>3</td>\n      <td>1</td>\n      <td>NOKIA CORP</td>\n      <td>23782051</td>\n      <td>NOKIA CORPORATION</td>\n      <td>2</td>\n      <td>COMPANY</td>\n      <td>2125445</td>\n      <td>NOKIA CORP</td>\n      <td>2</td>\n      <td>COMPANY</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>128</td>\n      <td>Nokia Siemens Networks Oy</td>\n      <td>Nokia Siemens Networks Oy</td>\n      <td>Karaportti 3,02610 Espoo</td>\n      <td>FI</td>\n      <td>FI1B1</td>\n      <td>3</td>\n      <td>112</td>\n      <td>NOKIA SIEMENS NETWORKS OY</td>\n      <td>23782129</td>\n      <td>NOKIA NETWORKS</td>\n      <td>2</td>\n      <td>COMPANY</td>\n      <td>2125445</td>\n      <td>NOKIA CORP</td>\n      <td>2</td>\n      <td>COMPANY</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>5217785</td>\n      <td>Nokia Corporation</td>\n      <td>Nokia Corporation</td>\n      <td>Espoo</td>\n      <td>FI</td>\n      <td>FI</td>\n      <td>0</td>\n      <td>1</td>\n      <td>NOKIA CORP</td>\n      <td>23782051</td>\n      <td>NOKIA CORPORATION</td>\n      <td>2</td>\n      <td>COMPANY</td>\n      <td>2125445</td>\n      <td>NOKIA CORP</td>\n      <td>2</td>\n      <td>COMPANY</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>5217811</td>\n      <td>Nokia Corporation</td>\n      <td>Nokia Corporation</td>\n      <td>NaN</td>\n      <td>FI</td>\n      <td>FI</td>\n      <td>0</td>\n      <td>1</td>\n      <td>NOKIA CORP</td>\n      <td>23782051</td>\n      <td>NOKIA CORPORATION</td>\n      <td>2</td>\n      <td>COMPANY</td>\n      <td>2125445</td>\n      <td>NOKIA CORP</td>\n      <td>2</td>\n      <td>COMPANY</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>5232170</td>\n      <td>Nokia Siemens Networks Oy</td>\n      <td>Nokia Siemens Networks Oy</td>\n      <td>Espoo</td>\n      <td>FI</td>\n      <td>FI</td>\n      <td>0</td>\n      <td>112</td>\n      <td>NOKIA SIEMENS NETWORKS OY</td>\n      <td>23782129</td>\n      <td>NOKIA NETWORKS</td>\n      <td>2</td>\n      <td>COMPANY</td>\n      <td>2125445</td>\n      <td>NOKIA CORP</td>\n      <td>2</td>\n      <td>COMPANY</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    </tr>\n    <tr>\n      <th>112235</th>\n      <td>85719932</td>\n      <td>VIKSTREM, Erik</td>\n      <td>ВИКСТРЁМ, Эрик</td>\n      <td>NaN</td>\n      <td>SE</td>\n      <td>SE</td>\n      <td>0</td>\n      <td>38919340</td>\n      <td>VIKSTREM ERIK</td>\n      <td>185719932</td>\n      <td>VIKSTREM, Erik</td>\n      <td>0</td>\n      <td>UNKNOWN</td>\n      <td>185719932</td>\n      <td>VIKSTREM, Erik</td>\n      <td>0</td>\n      <td>UNKNOWN</td>\n    </tr>\n    <tr>\n      <th>112236</th>\n      <td>85720336</td>\n      <td>HWANG, LING-CHI</td>\n      <td>HWANG, LING-CHI</td>\n      <td>NaN</td>\n      <td>TW</td>\n      <td>NaN</td>\n      <td>9</td>\n      <td>35599384</td>\n      <td>HWANG LING-CHI</td>\n      <td>185720336</td>\n      <td>HWANG, LING-CHI</td>\n      <td>0</td>\n      <td>UNKNOWN</td>\n      <td>185720336</td>\n      <td>HWANG, LING-CHI</td>\n      <td>0</td>\n      <td>UNKNOWN</td>\n    </tr>\n    <tr>\n      <th>112237</th>\n      <td>85720376</td>\n      <td>LI, I Chan</td>\n      <td>LI, I Chan</td>\n      <td>NaN</td>\n      <td>TW</td>\n      <td>NaN</td>\n      <td>9</td>\n      <td>38707281</td>\n      <td>LI I CHAN</td>\n      <td>185720376</td>\n      <td>LI, I Chan</td>\n      <td>0</td>\n      <td>UNKNOWN</td>\n      <td>185720376</td>\n      <td>LI, I Chan</td>\n      <td>0</td>\n      <td>UNKNOWN</td>\n    </tr>\n    <tr>\n      <th>112238</th>\n      <td>85720469</td>\n      <td>TING, Chia Ching</td>\n      <td>TING, Chia Ching</td>\n      <td>TW</td>\n      <td>TW</td>\n      <td>NaN</td>\n      <td>9</td>\n      <td>23937900</td>\n      <td>TING CHIA CHING</td>\n      <td>185720469</td>\n      <td>TING, Chia Ching</td>\n      <td>0</td>\n      <td>UNKNOWN</td>\n      <td>185720469</td>\n      <td>TING, Chia Ching</td>\n      <td>0</td>\n      <td>UNKNOWN</td>\n    </tr>\n    <tr>\n      <th>112239</th>\n      <td>85720500</td>\n      <td>WANG, YU-CHEIH</td>\n      <td>WANG, YU-CHEIH</td>\n      <td>NaN</td>\n      <td>TW</td>\n      <td>NaN</td>\n      <td>9</td>\n      <td>38204835</td>\n      <td>WANG YU-CHEIH</td>\n      <td>185720500</td>\n      <td>WANG, YU-CHEIH</td>\n      <td>0</td>\n      <td>UNKNOWN</td>\n      <td>185720500</td>\n      <td>WANG, YU-CHEIH</td>\n      <td>0</td>\n      <td>UNKNOWN</td>\n    </tr>\n  </tbody>\n</table>\n<p>112240 rows × 17 columns</p>\n</div>"
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pers_sector_primary = pers.groupby(\"han_id\", as_index=False)[\"psn_sector\"].agg(\n",
    "    lambda x: pd.Series.mode(x)[0]).rename(columns={\"psn_sector\":\"psn_sector_primary\"})\n",
    "persn = pers.merge(pers_sector_primary, on='han_id')\n",
    "persn"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "outputs": [
    {
     "data": {
      "text/plain": "           han_id         psn_sector_primary\n0             264  GOV NON-PROFIT UNIVERSITY\n1             627                    COMPANY\n2             974                    COMPANY\n3            1480                    COMPANY\n4            1699                    COMPANY\n...           ...                        ...\n106154  185719932                    UNKNOWN\n106155  185720336                    UNKNOWN\n106156  185720376                    UNKNOWN\n106157  185720469                    UNKNOWN\n106158  185720500                    UNKNOWN\n\n[106159 rows x 2 columns]",
      "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>han_id</th>\n      <th>psn_sector_primary</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>264</td>\n      <td>GOV NON-PROFIT UNIVERSITY</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>627</td>\n      <td>COMPANY</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>974</td>\n      <td>COMPANY</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>1480</td>\n      <td>COMPANY</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>1699</td>\n      <td>COMPANY</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>106154</th>\n      <td>185719932</td>\n      <td>UNKNOWN</td>\n    </tr>\n    <tr>\n      <th>106155</th>\n      <td>185720336</td>\n      <td>UNKNOWN</td>\n    </tr>\n    <tr>\n      <th>106156</th>\n      <td>185720376</td>\n      <td>UNKNOWN</td>\n    </tr>\n    <tr>\n      <th>106157</th>\n      <td>185720469</td>\n      <td>UNKNOWN</td>\n    </tr>\n    <tr>\n      <th>106158</th>\n      <td>185720500</td>\n      <td>UNKNOWN</td>\n    </tr>\n  </tbody>\n</table>\n<p>106159 rows × 2 columns</p>\n</div>"
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pers_sector_primary"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "outputs": [],
   "source": [
    "appln_merge = appln.merge(appln_title, on=\"appln_id\")#.merge(appln_pers,on=\"appln_id\")\n",
    "appln_merge.to_excel(\"appln_data.xlsx\", index=False)"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "outputs": [],
   "source": [
    "person_merge = appln_pers.merge(pers,on=\"person_id\")\n",
    "person_merge.to_excel(\"person_data.xlsx\", index=False)"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "outputs": [
    {
     "data": {
      "text/plain": "array(['FI', 'NL', 'FR', 'DE', 'DK', 'AT', 'SE', 'BE', 'TW', 'LU', 'CN',\n       'IT', 'HU', 'IE', 'SI', 'CZ', 'ES', 'HK', 'PL', 'CY', 'SK', 'PT',\n       'LT', 'EE', 'MT', 'GR', 'RO', 'BG', 'HR', 'MO', 'LV'], dtype=object)"
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pers[\"person_ctry_code\"].unique()"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "'cry_code'",
     "output_type": "error",
     "traceback": [
      "\u001B[1;31m---------------------------------------------------------------------------\u001B[0m",
      "\u001B[1;31mKeyError\u001B[0m                                  Traceback (most recent call last)",
      "File \u001B[1;32m~\\.conda\\envs\\MOME_BIGDATA\\lib\\site-packages\\pandas\\core\\indexes\\base.py:3649\u001B[0m, in \u001B[0;36mIndex.get_loc\u001B[1;34m(self, key)\u001B[0m\n\u001B[0;32m   3648\u001B[0m \u001B[38;5;28;01mtry\u001B[39;00m:\n\u001B[1;32m-> 3649\u001B[0m     \u001B[38;5;28;01mreturn\u001B[39;00m \u001B[38;5;28;43mself\u001B[39;49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43m_engine\u001B[49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43mget_loc\u001B[49m\u001B[43m(\u001B[49m\u001B[43mcasted_key\u001B[49m\u001B[43m)\u001B[49m\n\u001B[0;32m   3650\u001B[0m \u001B[38;5;28;01mexcept\u001B[39;00m \u001B[38;5;167;01mKeyError\u001B[39;00m \u001B[38;5;28;01mas\u001B[39;00m err:\n",
      "File \u001B[1;32m~\\.conda\\envs\\MOME_BIGDATA\\lib\\site-packages\\pandas\\_libs\\index.pyx:147\u001B[0m, in \u001B[0;36mpandas._libs.index.IndexEngine.get_loc\u001B[1;34m()\u001B[0m\n",
      "File \u001B[1;32m~\\.conda\\envs\\MOME_BIGDATA\\lib\\site-packages\\pandas\\_libs\\index.pyx:176\u001B[0m, in \u001B[0;36mpandas._libs.index.IndexEngine.get_loc\u001B[1;34m()\u001B[0m\n",
      "File \u001B[1;32mpandas\\_libs\\hashtable_class_helper.pxi:7080\u001B[0m, in \u001B[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001B[1;34m()\u001B[0m\n",
      "File \u001B[1;32mpandas\\_libs\\hashtable_class_helper.pxi:7088\u001B[0m, in \u001B[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001B[1;34m()\u001B[0m\n",
      "\u001B[1;31mKeyError\u001B[0m: 'cry_code'",
      "\nThe above exception was the direct cause of the following exception:\n",
      "\u001B[1;31mKeyError\u001B[0m                                  Traceback (most recent call last)",
      "Cell \u001B[1;32mIn[16], line 1\u001B[0m\n\u001B[1;32m----> 1\u001B[0m \u001B[43mperson_merge\u001B[49m\u001B[43m[\u001B[49m\u001B[38;5;124;43m\"\u001B[39;49m\u001B[38;5;124;43mcry_code\u001B[39;49m\u001B[38;5;124;43m\"\u001B[39;49m\u001B[43m]\u001B[49m\u001B[38;5;241m.\u001B[39munique()\n",
      "File \u001B[1;32m~\\.conda\\envs\\MOME_BIGDATA\\lib\\site-packages\\pandas\\core\\frame.py:3745\u001B[0m, in \u001B[0;36mDataFrame.__getitem__\u001B[1;34m(self, key)\u001B[0m\n\u001B[0;32m   3743\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mcolumns\u001B[38;5;241m.\u001B[39mnlevels \u001B[38;5;241m>\u001B[39m \u001B[38;5;241m1\u001B[39m:\n\u001B[0;32m   3744\u001B[0m     \u001B[38;5;28;01mreturn\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39m_getitem_multilevel(key)\n\u001B[1;32m-> 3745\u001B[0m indexer \u001B[38;5;241m=\u001B[39m \u001B[38;5;28;43mself\u001B[39;49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43mcolumns\u001B[49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43mget_loc\u001B[49m\u001B[43m(\u001B[49m\u001B[43mkey\u001B[49m\u001B[43m)\u001B[49m\n\u001B[0;32m   3746\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m is_integer(indexer):\n\u001B[0;32m   3747\u001B[0m     indexer \u001B[38;5;241m=\u001B[39m [indexer]\n",
      "File \u001B[1;32m~\\.conda\\envs\\MOME_BIGDATA\\lib\\site-packages\\pandas\\core\\indexes\\base.py:3651\u001B[0m, in \u001B[0;36mIndex.get_loc\u001B[1;34m(self, key)\u001B[0m\n\u001B[0;32m   3649\u001B[0m     \u001B[38;5;28;01mreturn\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39m_engine\u001B[38;5;241m.\u001B[39mget_loc(casted_key)\n\u001B[0;32m   3650\u001B[0m \u001B[38;5;28;01mexcept\u001B[39;00m \u001B[38;5;167;01mKeyError\u001B[39;00m \u001B[38;5;28;01mas\u001B[39;00m err:\n\u001B[1;32m-> 3651\u001B[0m     \u001B[38;5;28;01mraise\u001B[39;00m \u001B[38;5;167;01mKeyError\u001B[39;00m(key) \u001B[38;5;28;01mfrom\u001B[39;00m \u001B[38;5;21;01merr\u001B[39;00m\n\u001B[0;32m   3652\u001B[0m \u001B[38;5;28;01mexcept\u001B[39;00m \u001B[38;5;167;01mTypeError\u001B[39;00m:\n\u001B[0;32m   3653\u001B[0m     \u001B[38;5;66;03m# If we have a listlike key, _check_indexing_error will raise\u001B[39;00m\n\u001B[0;32m   3654\u001B[0m     \u001B[38;5;66;03m#  InvalidIndexError. Otherwise we fall through and re-raise\u001B[39;00m\n\u001B[0;32m   3655\u001B[0m     \u001B[38;5;66;03m#  the TypeError.\u001B[39;00m\n\u001B[0;32m   3656\u001B[0m     \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39m_check_indexing_error(key)\n",
      "\u001B[1;31mKeyError\u001B[0m: 'cry_code'"
     ]
    }
   ],
   "source": [
    "person_merge[\"cry_code\"].unique()"
   ],
   "metadata": {
    "collapsed": false
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}