RecSys ’18: HOP-Rec: High-Order Proximity for Implicit Recommendation

Recommendation system paper challenge (14/50)

paper link

What problem do they solve?

1. Recommend good item to users

What model do they propose?

HOP-rec: high-order proximity for implicit recommendation

The basic idea is combining matrix factorization (MF), (factorizing the observed direct interactions between users and items) and Graph-based models, (extracts indirect preferences from the graphs constructed by user-item interactions).

For graph model part, they applied random walk (RW) to approximate the high-order probabilistic matrix factorization with confidence weighting C(K)

For factorization part, they applied pairwise logistic loss

Data

Baseline Methods

Matrix Factorization (MF)

Bayesian Personalized Ranking (BPR)

Weighted Approximate-Rank Pairwise (WARP) loss

K-Order Statistic (K-OS) loss

Popularity-based Re-ranking (RP3 (β))

Result

Other related blogs:

Beyond Clicks: Dwell Time for Personalization

RecSys’15: Context-Aware Event Recommendation in Event-based Social Networks

RecSys16: Adaptive, Personalized Diversity for Visual Discovery

RecSys ’16: Local Item-Item Models for Top-N Recommendation

COLING’14: Deep Convolutional Neural Networks for Sentiment Analysis of Short Texts

NAACL’19: Utilizing BERT for Aspect-Based Sentiment Analysis via Constructing Auxiliary Sentence

RecSys ’17: Translation-based Recommendation

RecSys ’17: Modeling the Assimilation-Contrast Effects in Online Product Rating Systems: Debiasing and Recommendations

Best paper in RecSys:

https://recsys.acm.org/best-papers/

My Website:

https://light0617.github.io/#/