Reducing Popularity Bias in Recommender Systems through AUC-Optimal Negative Sampling.
Bin LiuErjia ChenBang WangPublished in: CoRR (2023)
Keyphrases
- recommender systems
- collaborative filtering
- positive and negative
- optimal solution
- dynamic programming
- logistic regression
- decision trees
- sampling algorithm
- user modeling
- error rate
- average reward reinforcement learning
- variance reduction
- random sampling
- monte carlo
- sample size
- classification accuracy
- user preferences
- matrix factorization
- naive bayes
- context aware
- receiver operating characteristic
- social media
- machine learning