Blurring-Sharpening Process Models for Collaborative Filtering

17 Nov 2022  ยท  Jeongwhan Choi, Seoyoung Hong, Noseong Park, Sung-Bae Cho ยท

Collaborative filtering is one of the most fundamental topics for recommender systems. Various methods have been proposed for collaborative filtering, ranging from matrix factorization to graph convolutional methods. Being inspired by recent successes of graph filtering-based methods and score-based generative models (SGMs), we present a novel concept of blurring-sharpening process model (BSPM). SGMs and BSPMs share the same processing philosophy that new information can be discovered (e.g., new images are generated in the case of SGMs) while original information is first perturbed and then recovered to its original form. However, SGMs and our BSPMs deal with different types of information, and their optimal perturbation and recovery processes have fundamental discrepancies. Therefore, our BSPMs have different forms from SGMs. In addition, our concept not only theoretically subsumes many existing collaborative filtering models but also outperforms them in terms of Recall and NDCG in the three benchmark datasets, Gowalla, Yelp2018, and Amazon-book. In addition, the processing time of our method is comparable to other fast baselines. Our proposed concept has much potential in the future to be enhanced by designing better blurring (i.e., perturbation) and sharpening (i.e., recovery) processes than what we use in this paper.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Recommendation Systems Amazon-Book BSPM-EM Recall@20 0.0733 # 3
nDCG@20 0.0609 # 4
Recommendation Systems Amazon-Book BSPM-LM Recall@20 0.0733 # 3
nDCG@20 0.0610 # 3
Collaborative Filtering Amazon-Book BSPM-LM Recall@20 0.0733 # 1
NDCG@20 0.0610 # 1
Collaborative Filtering Amazon-Book BSPM-EM Recall@20 0.0733 # 1
NDCG@20 0.0609 # 2
Recommendation Systems Gowalla BSPM-EM Recall@20 0.1920 # 1
nDCG@20 0.1597 # 3
Collaborative Filtering Gowalla BSPM-LM Recall@20 0.1901 # 3
NDCG@20 0.1570 # 5
Collaborative Filtering Gowalla BSPM-EM Recall@20 0.1920 # 1
NDCG@20 0.1597 # 2
Recommendation Systems Gowalla BSPM-LM Recall@20 0.1901 # 3
nDCG@20 0.157 # 7
Recommendation Systems Yelp2018 BSPM-EM Recall@20 0.0720 # 2
NDCG@20 0.0593 # 2
Collaborative Filtering Yelp2018 BSPM-EM NDCG@20 0.0593 # 2
Recall@20 0.0720 # 2
Collaborative Filtering Yelp2018 BSPM-LM NDCG@20 0.0584 # 3
Recall@20 0.0713 # 3
Recommendation Systems Yelp2018 BSPM-LM Recall@20 0.0713 # 3
NDCG@20 0.0584 # 3

Methods


No methods listed for this paper. Add relevant methods here