Abstract:Studying recommender systems with implicit feedback has become increasingly important. However, most existing works are designed in an offline setting while online recommendation is quite challenging due to the one-class nature of implicit feedback. In this paper, we propose an online collaborative filtering method for implicit feedback. We highlight three critical issues of existing works. First, when positive feedback arrives sequentially, if we treat all the other missing items for this given user as the negative samples, the mis-classified items will incur a large deviation since some items might appear as the positive feedback in the subsequent rounds. Second, the cost of missing a positive feedback should be much higher than that of having a false-positive. Third, the existing works usually assume that a fixed model is given prior to the learning task, which could result in poor performance if the chosen model is inappropriate. To address these issues, we propose a unified framework for Online Collaborative Filtering with Implicit Feedback (OCFIF). Motivated by the regret aversion, we propose a divestiture loss to heal the bias derived from the past mis-classified negative samples. Furthermore, we adopt cost-sensitive learning method to efficiently optimize the implicit MF model without imposing a heuristic weight restriction on missing data. By leveraging meta-learning, we dynamically explore a pool of multiple models to avoid the limitations of a single fixed model so as to remedy the drawback of manual/heuristic model selection. We also analyze the theoretical bounds of the proposed OCFIF method and conduct extensive experiments to evaluate its empirical performance on real-world datasets.

User Preference Modeling on Heterogeneous Implicit Feedback with Transfer Learning<sup>1</sup>

Confidence-Learning Based Collaborative Filtering With Heterogeneous Implicit Feedbacks

A Novel Confidence Estimation Method for Heterogeneous Implicit Feedback

Brand Recommendation Leveraging Heterogeneous Implicit Feedbacks

A Novel Framework To Process The Quantity And Quality Of User Behavior Data In Recommender Systems

Preference Transfer Model in Collaborative Filtering for Implicit Data

Latent Factor Transition for Dynamic Collaborative Filtering.

Personalized Recommendation with Hybrid Feedback by Refining Implicit Data

Improving Implicit Feedback-Based Recommendation through Multi-Behavior Alignment

Online Collaborative Filtering With Implicit Feedback

Modeling Dynamic Missingness of Implicit Feedback for Sequential Recommendation

Learning Transferrable Parameters for Long-tailed Sequential User Behavior Modeling

Learning Recommender Systems with Implicit Feedback Via Soft Target Enhancement

Hybrid Interest Modeling for Long-tailed Users

Learning and Transferring Social and Item Visibilities for Personalized Recommendation

A Hybrid Fuzzy Profiling-Nonnegative Latent Factor Model Considering Consumer Preference At Different Levels

Modeling Dynamic Missingness of Implicit Feedback for Recommendation.

Modeling User Purchase Preference Based on Implicit Feedback

Adaptive Pairwise Preference Learning for Collaborative Recommendation with Implicit Feedbacks.

Separating and Learning Latent Confounders to Enhancing User Preferences Modeling

Criterion-based Heterogeneous Collaborative Filtering for Multi-behavior Implicit Recommendation