dskg-construction/final_csv_files_transformation_dcat.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Make final OpenAIRE-csv file for transformation into RDF\n",
    "#Adds enriched information to csv file\n",
    "df = pd.read_csv(\"/OpenAire_classified.csv\")\n",
    "df_dskg = pd.read_csv(\"/DSKG_BETA_IDs.csv\", converters={\"contactPointID\":str})\n",
    "\n",
    "df[\"creatorPersonID\"] = \"\"\n",
    "df[\"creatorOrganizationID\"] = \"\"\n",
    "df[\"creatorAgentID\"] = \"\"\n",
    "df[\"publisherOrganizationID\"] = \"\"\n",
    "df[\"publisherAgentID\"] = \"\"\n",
    "df[\"contributorOrganizationID\"] = \"\"\n",
    "df[\"contributorAgentID\"] = \"\"\n",
    "df[\"contactPointID\"] = \"\"\n",
    "df[\"MAG_Disciplines_Absolute\"] = \"\"\n",
    "\n",
    "i = 0\n",
    "while i < len(df):\n",
    "    df[\"creatorPersonID\"][i] = df_dskg[\"creatorPersonID\"][i]\n",
    "    df[\"creatorOrganizationID\"][i] = df_dskg[\"creatorOrganizationID\"][i]\n",
    "    df[\"creatorAgentID\"][i] = df_dskg[\"creatorAgentID\"][i]\n",
    "    df[\"publisherOrganizationID\"][i] = df_dskg[\"publisherOrganizationID\"][i]\n",
    "    df[\"publisherAgentID\"][i] = df_dskg[\"publisherAgentID\"][i]\n",
    "    df[\"contributorOrganizationID\"][i] = df_dskg[\"contributorOrganizationID\"][i]\n",
    "    df[\"contributorAgentID\"][i] = df_dskg[\"contributorAgentID\"][i]\n",
    "    df[\"contactPointID\"][i] = df_dskg[\"contactPointID\"][i]\n",
    "    df[\"MAG_Disciplines_Absolute\"][i] = df_dskg[\"MAG_Disciplines_Absolute\"][i]\n",
    "    i += 1\n",
    "    \n",
    "df.to_csv(\"/OpenAire_Dataset_Transformation.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Make final Wikidata-csv for transformation into RDF\n",
    "df = pd.read_csv(\"/Wikidata_classified.csv\")\n",
    "df_dskg = pd.read_csv(\"/DSKG_BETA_IDs.csv\", converters={\"contactPointID\":str})\n",
    "\n",
    "df[\"creatorPersonID\"] = \"\"\n",
    "df[\"creatorOrganizationID\"] = \"\"\n",
    "df[\"creatorAgentID\"] = \"\"\n",
    "df[\"publisherOrganizationID\"] = \"\"\n",
    "df[\"publisherAgentID\"] = \"\"\n",
    "df[\"contributorOrganizationID\"] = \"\"\n",
    "df[\"contributorAgentID\"] = \"\"\n",
    "df[\"contactPointID\"] = \"\"\n",
    "df[\"MAG_Disciplines_Absolute\"] = \"\"\n",
    "\n",
    "i = 0\n",
    "while i < len(df):\n",
    "    df[\"creatorPersonID\"][i] = df_dskg[\"creatorPersonID\"][i + 588]\n",
    "    df[\"creatorOrganizationID\"][i] = df_dskg[\"creatorOrganizationID\"][i + 588]\n",
    "    df[\"creatorAgentID\"][i] = df_dskg[\"creatorAgentID\"][i + 588]\n",
    "    df[\"publisherOrganizationID\"][i] = df_dskg[\"publisherOrganizationID\"][i + 588]\n",
    "    df[\"publisherAgentID\"][i] = df_dskg[\"publisherAgentID\"][i + 588]\n",
    "    df[\"contributorOrganizationID\"][i] = df_dskg[\"contributorOrganizationID\"][i + 588]\n",
    "    df[\"contributorAgentID\"][i] = df_dskg[\"contributorAgentID\"][i + 588]\n",
    "    df[\"contactPointID\"][i] = df_dskg[\"contactPointID\"][i + 588]\n",
    "    df[\"MAG_Disciplines_Absolute\"][i] = df_dskg[\"MAG_Disciplines_Absolute\"][i + 588]\n",
    "    i += 1\n",
    "    \n",
    "df.to_csv(\"/Wikidata_Dataset_Transformation.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Make final CreatorPerson-csv for transformation into RDF\n",
    "df = pd.read_csv(\"/DSKG_BETA_IDs.csv\")\n",
    "\n",
    "new_dict = {}\n",
    "i = 0\n",
    "while i < len(df):\n",
    "    creator_ids = df[\"creatorPersonID\"][i]\n",
    "    creator_ids_split = str(creator_ids).split(\", \")\n",
    "    creator_names = df[\"creatorPersonName\"][i]\n",
    "    creator_names_split = str(creator_names).split(\", \") \n",
    "    \n",
    "    for cid, cn in zip(creator_ids_split, creator_names_split):\n",
    "        if str(cid) != \"nan\" and str(cn) != \"nan\":\n",
    "            new_dict[cid] = cn\n",
    "        \n",
    "    i += 1\n",
    "    \n",
    "df_creator_person = pd.DataFrame(index=range(0,len(new_dict)))\n",
    "df_creator_person[\"creatorPersonID\"] = \"\"\n",
    "df_creator_person[\"creatorPersonName\"] = \"\"\n",
    "df_creator_person[\"ORCID\"] = \"\"\n",
    "    \n",
    "counter = 0\n",
    "for key in new_dict:\n",
    "    df_creator_person[\"creatorPersonID\"][counter] = key\n",
    "    df_creator_person[\"creatorPersonName\"][counter] = new_dict[key]\n",
    "    counter += 1\n",
    "\n",
    "df_creator_person.to_csv(\"/DSKG_CreatorPerson_Transformation.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Creator_Organization\n",
    "df = pd.read_csv(\"/DSKG_BETA_IDs.csv\")\n",
    "\n",
    "new_dict = {}\n",
    "i = 0\n",
    "while i < len(df):\n",
    "    creator_ids = df[\"creatorOrganizationID\"][i]\n",
    "    creator_ids_split = str(creator_ids).split(\", \")\n",
    "    creator_names = df[\"creatorOrganizationName\"][i]\n",
    "    creator_names_split = str(creator_names).split(\", \") \n",
    "    \n",
    "    for cid, cn in zip(creator_ids_split, creator_names_split):\n",
    "        if str(cid) != \"nan\" and str(cn) != \"nan\":\n",
    "            new_dict[cid] = cn\n",
    "        \n",
    "    i += 1\n",
    "\n",
    "df_creator_person = pd.DataFrame(index=range(0,len(new_dict)))\n",
    "df_creator_person[\"creatorOrganizationID\"] = \"\"\n",
    "df_creator_person[\"creatorOrganizationName\"] = \"\"\n",
    "\n",
    "counter = 0\n",
    "for key in new_dict:\n",
    "    df_creator_person[\"creatorOrganizationID\"][counter] = key\n",
    "    df_creator_person[\"creatorOrganizationName\"][counter] = new_dict[key]\n",
    "    counter += 1\n",
    "    \n",
    "df_creator_person.to_csv(\"/DSKG_CreatorOrganization_Transformation.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Creator_Agent\n",
    "df = pd.read_csv(\"/DSKG_BETA_IDs.csv\")\n",
    "\n",
    "new_dict = {}\n",
    "i = 0\n",
    "while i < len(df):\n",
    "    creator_ids = df[\"creatorAgentID\"][i]\n",
    "    creator_ids_split = str(creator_ids).split(\", \")\n",
    "    creator_names = df[\"creatorAgentName\"][i]\n",
    "    creator_names_split = str(creator_names).split(\", \") \n",
    "    \n",
    "    for cid, cn in zip(creator_ids_split, creator_names_split):\n",
    "        if str(cid) != \"nan\" and str(cn) != \"nan\":\n",
    "            new_dict[cid] = cn\n",
    "        \n",
    "    i += 1\n",
    "\n",
    "df_creator_person = pd.DataFrame(index=range(0,len(new_dict)))\n",
    "df_creator_person[\"creatorAgentID\"] = \"\"\n",
    "df_creator_person[\"creatorAgentName\"] = \"\"\n",
    "\n",
    "counter = 0\n",
    "for key in new_dict:\n",
    "    df_creator_person[\"creatorAgentID\"][counter] = key\n",
    "    df_creator_person[\"creatorAgentName\"][counter] = new_dict[key]\n",
    "    counter += 1\n",
    "    \n",
    "df_creator_person.to_csv(\"/DSKG_CreatorAgent_Transformation.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#publisher_Organization\n",
    "df = pd.read_csv(\"/DSKG_BETA_IDs.csv\")\n",
    "\n",
    "new_dict = {}\n",
    "i = 0\n",
    "while i < len(df):\n",
    "    creator_ids = df[\"publisherOrganizationID\"][i]\n",
    "    creator_ids_split = str(creator_ids).split(\", \")\n",
    "    creator_names = df[\"publisherOrganizationName\"][i]\n",
    "    creator_names_split = str(creator_names).split(\", \") \n",
    "    \n",
    "    for cid, cn in zip(creator_ids_split, creator_names_split):\n",
    "        if str(cid) != \"nan\" and str(cn) != \"nan\":\n",
    "            new_dict[cid] = cn\n",
    "        \n",
    "    i += 1\n",
    "\n",
    "df_creator_person = pd.DataFrame(index=range(0,len(new_dict)))\n",
    "df_creator_person[\"publisherOrganizationID\"] = \"\"\n",
    "df_creator_person[\"publisherOrganizationName\"] = \"\"\n",
    "\n",
    "counter = 0\n",
    "for key in new_dict:\n",
    "    df_creator_person[\"publisherOrganizationID\"][counter] = key\n",
    "    df_creator_person[\"publisherOrganizationName\"][counter] = new_dict[key]\n",
    "    counter += 1\n",
    "\n",
    "df_creator_person.to_csv(\"/DSKG_PublisherOrganization_Transformation.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#publisher_Agent\n",
    "df = pd.read_csv(\"/DSKG_BETA_IDs.csv\")\n",
    "\n",
    "new_dict = {}\n",
    "i = 0\n",
    "while i < len(df):\n",
    "    creator_ids = df[\"publisherAgentID\"][i]\n",
    "    creator_ids_split = str(creator_ids).split(\", \")\n",
    "    creator_names = df[\"publisherAgentName\"][i]\n",
    "    creator_names_split = str(creator_names).split(\", \") \n",
    "    \n",
    "    for cid, cn in zip(creator_ids_split, creator_names_split):\n",
    "        if str(cid) != \"nan\" and str(cn) != \"nan\":\n",
    "            new_dict[cid] = cn\n",
    "        \n",
    "    i += 1\n",
    "\n",
    "df_creator_person = pd.DataFrame(index=range(0,len(new_dict)))\n",
    "df_creator_person[\"publisherAgentID\"] = \"\"\n",
    "df_creator_person[\"publisherAgentName\"] = \"\"\n",
    "\n",
    "counter = 0\n",
    "for key in new_dict:\n",
    "    df_creator_person[\"publisherAgentID\"][counter] = key\n",
    "    df_creator_person[\"publisherAgentName\"][counter] = new_dict[key]\n",
    "    counter += 1\n",
    "    \n",
    "df_creator_person.to_csv(\"/DSKG_PublisherAgent_Transformation.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#contributor_Organization\n",
    "df = pd.read_csv(\"/DSKG_BETA_IDs.csv\")\n",
    "\n",
    "new_dict = {}\n",
    "i = 0\n",
    "while i < len(df):\n",
    "    creator_ids = df[\"contributorOrganizationID\"][i]\n",
    "    creator_ids_split = str(creator_ids).split(\", \")\n",
    "    creator_names = df[\"contributorOrganizationName\"][i]\n",
    "    creator_names_split = str(creator_names).split(\", \") \n",
    "    \n",
    "    for cid, cn in zip(creator_ids_split, creator_names_split):\n",
    "        if str(cid) != \"nan\" and str(cn) != \"nan\":\n",
    "            new_dict[cid] = cn\n",
    "        \n",
    "    i += 1\n",
    "\n",
    "df_creator_person = pd.DataFrame(index=range(0,len(new_dict)))\n",
    "df_creator_person[\"contributorOrganizationID\"] = \"\"\n",
    "df_creator_person[\"contributorOrganizationName\"] = \"\"\n",
    "\n",
    "counter = 0\n",
    "for key in new_dict:\n",
    "    df_creator_person[\"contributorOrganizationID\"][counter] = key\n",
    "    df_creator_person[\"contributorOrganizationName\"][counter] = new_dict[key]\n",
    "    counter += 1\n",
    "    \n",
    "df_creator_person.to_csv(\"/DSKG_ContributorOrganization_Transformation.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#contributor_Agent\n",
    "import pandas as pd\n",
    "df = pd.read_csv(\"/DSKG_BETA_IDs.csv\")\n",
    "\n",
    "new_dict = {}\n",
    "i = 0\n",
    "while i < len(df):\n",
    "    creator_ids = df[\"contributorAgentID\"][i]\n",
    "    creator_ids_split = str(creator_ids).split(\", \")\n",
    "    creator_names = df[\"contributorAgentName\"][i]\n",
    "    creator_names_split = str(creator_names).split(\", \") \n",
    "    \n",
    "    for cid, cn in zip(creator_ids_split, creator_names_split):\n",
    "        if str(cid) != \"nan\" and str(cn) != \"nan\":\n",
    "            new_dict[cid] = cn\n",
    "        \n",
    "    i += 1\n",
    "\n",
    "df_creator_person = pd.DataFrame(index=range(0,len(new_dict)))\n",
    "df_creator_person[\"contributorAgentID\"] = \"\"\n",
    "df_creator_person[\"contributorAgentName\"] = \"\"\n",
    "\n",
    "counter = 0\n",
    "for key in new_dict:\n",
    "    df_creator_person[\"contributorAgentID\"][counter] = key\n",
    "    df_creator_person[\"contributorAgentName\"][counter] = new_dict[key]\n",
    "    counter += 1\n",
    "    \n",
    "df_creator_person.to_csv(\"/DSKG_ContributorAgent_Transformation.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#contactPoint\n",
    "df = pd.read_csv(\"/DSKG_BETA_IDs.csv\", converters={\"contactPointID\":str})\n",
    "\n",
    "new_dict = {}\n",
    "i = 0\n",
    "while i < len(df):\n",
    "    creator_ids = df[\"contactPointID\"][i]\n",
    "    creator_names = df[\"contactPointName\"][i]\n",
    "    if str(creator_ids) != \"nan\" and str(creator_names) != \"nan\":\n",
    "        new_dict[creator_ids] = creator_names\n",
    "        \n",
    "    i += 1\n",
    "\n",
    "df_creator_person = pd.DataFrame(index=range(0,len(new_dict)))\n",
    "df_creator_person[\"contactPointID\"] = \"\"\n",
    "df_creator_person[\"contactPoint\"] = \"\"\n",
    "df_creator_person[\"contactPointEmail\"] = \"\"\n",
    "\n",
    "counter = 0\n",
    "for key in new_dict:\n",
    "    df_creator_person[\"contactPointID\"][counter] = key\n",
    "    df_creator_person[\"contactPoint\"][counter] = new_dict[key]\n",
    "    counter += 1\n",
    "    \n",
    "df_creator_person.to_csv(\"DSKG_ContactPoint_Transformation.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}