search

o19s / skipchunk

Extracts a latent knowledge graph from text and index/query it in elasticsearch or solr
MIT License
19 stars 2 forks source link

Does content enrichment and querying belong here? #13

Closed binarymax closed 3 years ago

binarymax commented 4 years ago

Skipchunk started as a latent knowledge graph extractor and autocomplete library. It has grown to include general content and query enrichment with indexing and querying. Should these latter features be split out into a separate library or folded into the Hello-NLP service?

Architecture decisions need to be made soon.

binarymax commented 4 years ago

The answer: no, it doesn't. Not for querying the search index at least. Graph querying will stay in Skipchunk but the pass-through query and content enrichment for the search index is moving to Hello-NLP.

binarymax commented 4 years ago

Hello-NLP now supports a robust content enrichment pipeline, and query enrichment is almost done. Time to start ripping out the enrichment stuff from skipchunk and decoupling content indexing from this library.

binarymax commented 3 years ago

Done!