Session 09 Supplement
Combining AI and Knowledge Extraction to Solve Complex, Underspecified Problems
DRAFT: This content is under development. Check back closer to the course start date.
Learning Objectives
- Scope underspecified problems: from vague question to actionable plan
- Build pipelines combining embeddings, graphs, clustering, and LLMs
- Evaluate results without ground truth
- Apply techniques to a real case study (HMC conversational clustering)
Topics
- Scoping underspecified problems
- Combining techniques into pipelines
- Case study: HMC clustering problem
- Evaluating without ground truth
Materials
Slides, readings, and exercises will be posted here.