PRP builds pseudo-rankings by noise injection and trains recommenders with a ranking loss, improving top-K metrics over pairwise baselines on four datasets.
Y.; Chen, Y.; and Cong, G
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
From Pairwise to Ranking: Climbing the Ladder to Ideal Collaborative Filtering with Pseudo-Ranking
PRP builds pseudo-rankings by noise injection and trains recommenders with a ranking loss, improving top-K metrics over pairwise baselines on four datasets.