philippchr / wikidata-core-for-qa Goto Github PK

If I just have a big ttl file, what should I do with it? Is there any way to split this file into small ones, so that I can convert those small ttl files into nt files and use this code to filter them?

Hi, there.
Thank you for your code and it's clear and helpful. I'm trying to use it to process KB and have some questions.

• I'm trying to understand the filtering process. Have you compared the size of KB (predicates, entities, etc. ) before and after using your code? or how many predicates have your filtered out? I notice that you just deleted ExternalID and GeoShape for predicates and filtered out some entities using rules. Is that because the ExternalID and GeoShape takes a large proportion?
• Why resolving the qualifiers is helpful for question answering, I'm curious about some detailed effects of this step. I think maybe skipping this step is okay.
• Just my interest. As you know, images and videos need a lot of memory to store, but they don't exit in .ttl and .nt files (so for as i know). Do you know how these special types of KB information are stored .ttl or .nt files?

I appreciate it for your precious time.