data/processed/framing.json

{
  "project": "E-commerce Recommendation System",
  "dataset": "McAuley-Lab/Amazon-Reviews-2023",
  "subset": "raw_review_All_Beauty",
  "stats": {
    "n_users": 631986,
    "n_items": 112565,
    "n_interactions": 701528,
    "sparsity_pct": 99.999
  },
  "ml_task": "Top-K ranking recommendation",
  "primary_metric": "ndcg@10",
  "cold_start_strategy": "Popularity-based fallback",
  "business_kpis": {
    "primary": "Click-through rate (CTR) uplift vs baseline",
    "secondary": "Conversion rate on recommended items",
    "guardrail": "Recommendation diversity (avoid filter bubbles)"
  }
}