NII internship 2018

Research project for 2017 2nd Call NII International Internship.

Main target: Analysis of Social Network of seiyuu (anime voice actors).

This is my first research experience, please be kind and understanding. Hope this will be useful to someone. Thank you!

Folders

• Code: contains scripts used for getting information, making graphs and run predictions.
• Data: json and rdf data backup, may be outdated.
• Graphics: backup of graphics.
• PopularityPredictionResults: results backup for an execution of "popularity predictions" script.
• Presentation: latex files for presentation.
• Report: latex files for report.
• SocialNetworks: backup of some of the social networks built.

Steps to get RDF data

1. Get list of seiyuu from Wikidata.

   • Using: retrieveSeiyuuListFromWikidata.py
   • Parameters: name of output file, limit

2. Get seiyuu info and store it locally using mongodb in order to not depend on API limit.

   • Using: getSeiyuuInfo.py
   • Parameters: input file name, output file name, from seiyuu index, to seiyuu index, limit to x api consults

     Input file is the output from point 1

3. Convert seiyuu information to RDF-turtle and store it locally using virtuoso.

   • Using: seiyuuInfoToRDF.py
   • parameters: input file name, output file name

4. Get anime info and store it locally using mongodb in order to not depend on API limit.

   • Using: getAnimeInfo.py
   • Parameters: input file name, output file name, from anime index, to anime index, limit to x api consults

     Input file is actually a list of anime uris (jikan API page) this can be obtained by querying virtuoso server

5. Convert anime information to RDF-turtle and store it locally using virtuoso.

   • Using: animeInfoToRDF.py
   • Parameters: input file name, output file name

6. There are a couple of scripts that recover certain information, such as debut or personal data. This can be separately and added to local database if needed.