digiarchitect / recs

DNA-seeded expansion · RSW minimization · weighted multi-dim scoring · streamed live

🎬 FIND SIMILAR MOVIES

Pick a film. The pipeline walks the target’s DNA — directors, writers, cast, genres, keywords — to build a candidate pool, then uses RSW to minimize redundancy and rank by weighted role contribution. TMDB’s /similar and /recommendations are supplemental, not seed.

target → director films → writer films → cast films → /discover by genre → /discover by keyword → /similar + /recs (supplemental) → merge → enrich top 30 → RSW → top 8
combined = director×5 + writer×4.5 + cinematographer×3.5 + composer×3 + keyword×4 + cast×3 + collection×3 + overview×3 + genreCorr×2 + company×1.5 + genre×1.5 + cluster×1.5 + editor×1 + language×0.25 + country×0.25 + tmdb_bonus
Data: TMDB · Pipeline: SSE-streamed DNA expansion · /discover · RSW minimization · weighted scoring
REFERENCES
  1. Choi, S.-M., Ko, S.-K. & Han, Y.-S. (2012). A movie recommendation algorithm based on genre correlations. Expert Systems with Applications, 39(9), 8079–8085. doi:10.1016/j.eswa.2012.01.132
  2. Helali, M. et al. (2025). Optimization of movie recommender system using differential evolutionary bees algorithm for clustering. Progress in Artificial Intelligence, 14, 581–606. doi:10.1007/s13748-025-00378-x
  3. DACTAL — in-memory graph engine for relational similarity queries across crew filmographies.

  /\
 ( o.o)
  > ^ <
 /|   |\
(_|   |_)