Visualizing the Output of Your Search AI
summary
A search app combines Apache Solr/Fusion with a TensorFlow convolutional model applied in the index pipeline, using transfer learning from e‑commerce review sentiment and a bundled model zip served via Spark. Superset connects to Solr (via SQLAlchemy/Hive) to surface prediction facets and give an audit trail. The pipeline drives navigation and trading signals and empirically returns 48% return on $8,000 in a personal experiment, with TensorFlow model served via Spark and Superset dashboards for audit trail facilitating inspection and iteration.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:53The Search App
- 4:00Indexes vs Tensorflow
- 8:42The App
- 14:25The Superset
- 15:08Fusion Instance
- 16:03Lab
- 18:00Admin UI
- 22:25Protocol Buffer
- 23:35Query Workbench
- 26:37Superset
- 27:27Database
- 32:24Tables
- 32:57Charts
- 35:19Market Sentiment
- 36:06Dashboard
- 38:00Multiple Collections
- 45:38Superset vs Zeppelin
related talks
38mAI-Ready Data Foundation Using Amazon OpenSearch
38mA Visual AI Search Flow Builder for OpenSearch: Build GenAI and Vector...
46mDelivering the next wave of innovation with OpenSearch
36mBuilding Next-gen Search With OpenSearch
26mStreamlining Property Search Across Multiple Platforms
31mVue, You and AI Search
33mRelevant Data Analysis: Apache Solr Analytics #bbuzz
24mEmpowering Vue Apps with AI