Jun 28, 2019 · Artificial Intelligence

iQIYI's RSLIME: A Novel Feature Importance Analysis Method for Video Recommendation Systems

iQIYI introduces RSLIME, a model‑agnostic, sample‑level feature importance method for its three‑stage small‑video recommendation system, enabling interpretable analysis of a complex ranking module that combines DNN, GBDT, and FM, and demonstrating stable, AUC‑correlated insights for optimization and feature selection.

DNNFMGBDT

0 likes · 11 min read

iQIYI's RSLIME: A Novel Feature Importance Analysis Method for Video Recommendation Systems

Meituan Technology Team

Mar 18, 2016 · Artificial Intelligence

Why FM and FFM Still Dominate Large‑Scale Sparse CTR Prediction

This article explains the principles of Factorization Machines (FM) and Field‑aware Factorization Machines (FFM), their implementation details, and how Meituan‑Dianping applied FFM in a DSP platform to achieve superior CTR and CVR estimation for sparse, high‑dimensional advertising data.

AdvertisingCTR predictionDSP

0 likes · 4 min read

Why FM and FFM Still Dominate Large‑Scale Sparse CTR Prediction