Chatting with Solr
summary
Demonstrates using the Solr Tagger as a first-stage query interpreter: a JSON /tag endpoint plus a collection built with catenateGraphFilterFactory tags known entities (example: a 124,000-city list, dozens of “Berlin” hits reduced to one with a geo FQ). Shows a chatbot prototype “Lou” that replaces tagged tokens with slotted grammar queries, a grammar collection that maps types (city, color, brand) to actions, and production use inside Lucidworks Fusion where the tagger runs fast enough to be replicated to all nodes.
AI-written summary · about
chapters (7)
- 0:00Introduction and setup
- 3:24Talk overview and agenda
- 4:07Solar text tagger basics
- 16:14Live tagger demo
- 22:35Building chat with tagger
- 27:47Live chat demo
- 37:42Conclusion and Q&A
related talks
26mRevolutionizing Search: Unveiling Conversational Search
15mRAG Beyond Chatbots
31mSponsored Session: Building a Production Open Source RAG Application using LlamaIndex and Cassandra
41mFrom Chat to RAG to MCP: Enhancing Java Applications with AI
49mRAG: from dumb implementation to serious results
01h 02mBuilding full stack AI chatbots with Java - livecoding
20mAI-driven Unstructured Data Extraction Using Apache Camel and LangChain4J
44mThe RAG Really Ties the App Together