Be Causal: De-biasing Social Network Confounding in Recommendation

by   Qian Li, et al.
University of Technology Sydney

In recommendation systems, the existence of the missing-not-at-random (MNAR) problem results in the selection bias issue, degrading the recommendation performance ultimately. A common practice to address MNAR is to treat missing entries from the so-called "exposure" perspective, i.e., modeling how an item is exposed (provided) to a user. Most of the existing approaches use heuristic models or re-weighting strategy on observed ratings to mimic the missing-at-random setting. However, little research has been done to reveal how the ratings are missing from a causal perspective. To bridge the gap, we propose an unbiased and robust method called DENC (De-bias Network Confounding in Recommendation) inspired by confounder analysis in causal inference. In general, DENC provides a causal analysis on MNAR from both the inherent factors (e.g., latent user or item factors) and auxiliary network's perspective. Particularly, the proposed exposure model in DENC can control the social network confounder meanwhile preserves the observed exposure information. We also develop a deconfounding model through the balanced representation learning to retain the primary user and item features, which enables DENC generalize well on the rating prediction. Extensive experiments on three datasets validate that our proposed model outperforms the state-of-the-art baselines.


Deep Causal Reasoning for Recommendations

Traditional recommender systems aim to estimate a user's rating to an it...

Attentive Social Recommendation: Towards User And Item Diversities

Social recommendation system is to predict unobserved user-item rating v...

UniWalk: Explainable and Accurate Recommendation for Rating and Network Data

How can we leverage social network data and observed ratings to correctl...

Causal Embeddings for Recommendation: An Extended Abstract

Recommendations are commonly used to modify user's natural behavior, for...

From Intervention to Domain Transportation: A Novel Perspective to Optimize Recommendation

The interventional nature of recommendation has attracted increasing att...

"Click" Is Not Equal to "Like": Counterfactual Recommendation for Mitigating Clickbait Issue

Recommendation is a prevalent and critical service in information system...

Uncertainty Calibration for Counterfactual Propensity Estimation in Recommendation

In recommendation systems, a large portion of the ratings are missing du...

Please sign up or login with your details

Forgot password? Click here to reset