Completed
Building the Explainability Index: Relevancy via Labels & Models
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Making Recommendations Explainable
Automatically move to the next video in the Classroom when playback concludes
- 1 Welcome & What You’ll Learn: Making Recommendations Explainable
- 2 What “Explainable Recommendations” Mean User View
- 3 Why Change the System: User Reports & Developer Traceability
- 4 The Classic Recommender Pipeline Overview
- 5 Embeddings 101: Content/Collaborative Filtering Basics
- 6 User Profile Stage: Why Single Embeddings Are Hard to Explain
- 7 Candidate Selection with KNN/HNSW: Speed vs. Quality Issues
- 8 Ranking + Business Logic: Scoring and Diversifying the Feed
- 9 New Explainable Approach: Store Positive Publication IDs
- 10 Building the Explainability Index: Relevancy via Labels & Models
- 11 Ordering Candidates: Attractiveness CTR → Predictive Model
- 12 Generating the Explanation Text Users See
- 13 Keeping Positives Manageable: Diversity + Ranking Positives
- 14 Smarter Candidate Quotas with Bandit-Style Allocation
- 15 Wrap-Up: Transparency, Better Relevance, Better Debugging
