Finding answers faster with intelligent search
summary
Amazon Kendra replaces an Elasticsearch keyword approach for Reg Ranger, using NLU plus multiple ML models to return direct answers, passages and ranked documents; a proof of concept finds relevant results in 95% of test cases versus the old engine. Integration uses native connectors and direct calls to the batch PutDocument API, pulls FAQ/leading-practice content from RDS Postgres, goes live summer 2020, and estimates $100K one-time savings plus thousands monthly in reduced content labor.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 0:37The problem with enterprise search
- 2:29Natural language queries
- 9:13Use cases
- 11:50Regranger
- 18:59Kendra
related talks
28mGive users relevant answers quickly with intelligent search
30mFind Accurate Information Faster with Intelligent Search
26mEmpower your business use cases
29mSolve common business problems with AI and ML
15mMastering RAG and Vector Search with Elastic
47mLatest innovations from Elasticsearch: Drive speed, scale & relevance (PRT266)
35mBuilding an enterprise NLSE ...
26mAdvancing Multi-Modal Search Capabilities in Search Pipeline