egerber / spaCy-entity-linker

spaCy module for linking text to Wikidata items
MIT License
215 stars 32 forks source link

Reproducing the underlying SQLite database #22

Open dennlinger opened 1 year ago

dennlinger commented 1 year ago

Hi,
I was just looking through some of the TODO's in the README, and found that a general limitation is the available offline DB for querying relevant information. E.g., to obtain URLs for images associated with entities, this would require access to property 18, which is not currently included in the statements table.

Given that Wikidata is also constantly updating their knowledge base (e.g., "COVID-19" is not currently included), I was wondering if there is any chance @egerber still has the filtering script somewhere, which would allow updates to the database and subsequently allow for optimizations in a more general direction.

Best, Dennis