7/25
- Garrick:
- Testing plugin
- Looking at document prior stuff
- Query independent – researched and tested using client-side script (more complicated query)
- Need to store priors ahead of time
- Tested, works fine, is pretty easy
- Query dependent priors
- Requires an initial retrieval
- Looking at plugins
- "painless" hah.
- Have code that's suggestive
- Query independent – researched and tested using client-side script (more complicated query)
- Hasn't run document expansion stuff
- Mike:
- Move elastichsearch repo
- ES integration tests workingish
- Testing finalization/documentation
- Re-uploaded Indri/Maven artifacts and will push the final artifacts out
- Snapshot ir-utils, es plugin, biocaddie
- Craig
- Shutdown SDSC instances, move data
- Release
- Final documentation/report
7/18
- Contract ends 7/30
- What's left
- ElasticSearch plugin – move repo (Mike)
- Testing – at least a manual test plan, automated would be great (Mike)
- PubMed ingest process (Craig)
- biocaddie + plugin repo release (Craig)
- Collect all data in place
- Documentation/presentation
- Bonus
- Parallel documentation
- Kubernetes review
- Publish data?
- Doc expansion on OHSUMED + Genomics (Garrick)
- Also PubMed expansion (Craig)
- "Priors" – if we wanted to implement priors in Lucene/ElasticSearch, how would we?
...