This PR addresses majority of the tasks listed in #172
Changelog:
Aligned production and staging builds for graph and keyphrase service
Added new connections to graph which are formed from the highlight/summary info - #172
Tagging segments as highlight=True to query high-fidelity data. The non-highlight segments are not deleted as they could be useful for other tasks
New node and predicate called SummaryGroup which contains summary group informations - Keyphrases, Entities, Users grouped together. Will be helpful for Related Meetings
Added mind filtrations for all the minds except Customer support mind
Regex-based candidate keyphrase extraction
Filtering PER, LOC type entities from final summaries
Fixed Dgraph-specific dateTime format for indexing and querying purpose - #156
Other tasks merged in this PR include:
Entity and keyphrase filtration based on entity-graph
Using regex-pattern to get candidate phrases
Keyphrase and entity post-processing - #177
Other closed issues that have been further enhanced/tested in this PR are: #163, #162, #177, #183
This PR addresses majority of the tasks listed in #172 Changelog:
highlight=Trueto query high-fidelity data. The non-highlight segments are not deleted as they could be useful for other tasksSummaryGroupwhich contains summary group informations - Keyphrases, Entities, Users grouped together. Will be helpful forRelated MeetingsPER,LOCtype entities from final summariesdateTimeformat for indexing and querying purpose - #156Other tasks merged in this PR include:
Other closed issues that have been further enhanced/tested in this PR are: #163, #162, #177, #183