# Addressing Confounding Feature Issue for Causal Recommendation

arXiv: https://arxiv.org/abs/2205.06532 · HTML: https://mlsanity.com/paper/2205.06532/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 May 2022.

## Built on by (most cited first, 12 of 12)

- [Mixture of Experts (MoE): A Big Data Perspective](https://mlsanity.com/paper/2501.16352/) (2025)
- [Causality-Enhanced Behavior Sequence Modeling in LLMs for Personalized Recommendation](https://mlsanity.com/paper/2410.22809/) (2024)
- [Debiasing Recommendation by Learning Identifiable Latent Confounders](https://mlsanity.com/paper/2302.05052/) (2023)
- [Rethinking Missing Data: Aleatoric Uncertainty-Aware Recommendation](https://mlsanity.com/paper/2209.11679/) (2022)
- [Causal Disentanglement for Regulating Social Influence Bias in Social Recommendation](https://mlsanity.com/paper/2403.03578/) (2024)
- [Policy-Guided Causal State Representation for Offline Reinforcement Learning Recommendation](https://mlsanity.com/paper/2502.02327/) (2025)
- [Interaction-Data-guided Conditional Instrumental Variables for Debiasing Recommender Systems](https://mlsanity.com/paper/2408.09651/) (2024)
- [Causal Intervention for Fairness in Multi-behavior Recommendation](https://mlsanity.com/paper/2209.04589/) (2022)
- [Teleporter Theory: A General and Simple Approach for Modeling Cross-World Counterfactual Causality](https://mlsanity.com/paper/2406.11501/) (2024)
- [Advancements in Recommender Systems: A Comprehensive Analysis Based on Data, Algorithms, and Evaluation](https://mlsanity.com/paper/2407.18937/) (2024)
- [Modeling Endogenous Logic: Causal Neuro-Symbolic Reasoning Model for Explainable Multi-Behavior Recommendation](https://mlsanity.com/paper/2601.21335/) (2026)
- [Generative Pseudo-Labeling for Pre-Ranking with LLMs](https://mlsanity.com/paper/2602.20995/) (2026)

## Abstract (arXiv)

In recommender system, some feature directly affects whether an interaction would happen, making the happened interactions not necessarily indicate user preference. For instance, short videos are objectively easier to be finished even though the user does not like the video. We term such feature as confounding feature, and video length is a confounding feature in video recommendation. If we fit a model on such interaction data, just as done by most data-driven recommender systems, the model will be biased to recommend short videos more, and deviate from user actual requirement. This work formulates and addresses the problem from the causal perspective. Assuming there are some factors affecting both the confounding feature and other item features, e.g., the video creator, we find the confounding feature opens a backdoor path behind user item matching and introduces spurious correlation. To remove the effect of backdoor path, we propose a framework named Deconfounding Causal Recommendation (DCR), which performs intervened inference with do-calculus. Nevertheless, evaluating do calculus requires to sum over the prediction on all possible values of confounding feature, significantly increasing the time cost. To address the efficiency challenge, we further propose a mixture-of experts (MoE) model architecture, modeling each value of confounding feature with a separate expert module. Through this way, we retain the model expressiveness with few additional costs. We demonstrate DCR on the backbone model of neural factorization machine (NFM), showing that DCR leads to more accurate prediction of user preference with small inference time cost.