Adding Narrative to BI Dashboards with Natural Language...
summary
Claims that adding generated sentences to dashboards makes data actionable and speeds work by automating repetitive reports. Uses Accelerate Text (NLG) paired with Metabase (BI), an open source pipeline combining a rules-based + data-driven generation approach rather than pure neural nets. Shows production deployments: e commerce product descriptions at scale (hundreds of thousands of items), bank AML reporting, and operational reports for shipping and factories, and reports consistent, precise output that reduces manual writing.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:16Data vs story messaging
- 3:24Limitations of data viz
- 5:55Natural language domains
- 8:49Gap in BI dashboards
- 10:37Low BI tool adoption
- 12:02Adding narrative to dashboards
- 13:30Accelerated Text design
- 16:22Data science in NLG
- 18:14Why neural networks fail
- 20:03NLG in e-commerce
- 21:50Benefits of NLG systems
- 23:30Q&A
related talks
19mInsights for Everyone - Democratising Analytics
16mGen AI Killed BI. Claude Says It's Fake News!
20mNatural Language Interfaces for SaaS: Transforming Workflows with ConversationalAI
35mUnlocking Business Insights with Natural Language and Generative AI
03mSay Goodbye to the Old BI: AI and Business Intelligence
24mData Pipelines to Dashboards
33mOwn Your Story: Turning Generative AI into Human-Sized Narratives
17mUnlocking the Power of AI for Seamless Self-Service BI Transfer