CausalRec: A CausalBoost Attention Model for Sequential Recommendation

2025-10-24

AI Summary

Existing co-occurrence-based sequential recommendation methods are vulnerable to spurious correlations and struggle to capture the true underlying causes of user behavior. To address this, we propose a causal attention framework—the first to integrate causal inference into the attention mechanism for sequential recommendation. Our approach employs causal discovery algorithms to construct a user-behavior causal graph and introduces the CausalBooster module to perform causal calibration of attention weights, explicitly disentangling causal effects from statistical co-occurrences. Experiments on multiple real-world datasets demonstrate substantial improvements: average Hit Rate and NDCG increase by 7.21% and 8.65%, respectively. These results validate the effectiveness of causal modeling in enhancing both recommendation accuracy and interpretability.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Knowledge Representation and Reasoning: Action, Change, and Causality

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Recent advances in correlation-based sequential recommendation systems have demonstrated substantial success. Specifically, the attention-based model outperforms other RNN-based and Markov chains-based models by capturing both short- and long-term dependencies more effectively. However, solely focusing on item co-occurrences overlooks the underlying motivations behind user behaviors, leading to spurious correlations and potentially inaccurate recommendations. To address this limitation, we present a novel framework that integrates causal attention for sequential recommendation, CausalRec. It incorporates a causal discovery block and a CausalBooster. The causal discovery block learns the causal graph in user behavior sequences, and we provide a theory to guarantee the identifiability of the learned causal graph. The CausalBooster utilizes the discovered causal graph to refine the attention mechanism, prioritizing behaviors with causal significance. Experimental evaluations on real-world datasets indicate that CausalRec outperforms several state-of-the-art methods, with average improvements of 7.21% in Hit Rate (HR) and 8.65% in Normalized Discounted Cumulative Gain (NDCG). To the best of our knowledge, this is the first model to incorporate causality through the attention mechanism in sequential recommendation, demonstrating the value of causality in generating more accurate and reliable recommendations.

Problem

• Addresses spurious correlations in sequential recommendation systems
• Identifies causal relationships in user behavior sequences
• Enhances attention mechanisms using discovered causal graphs

Innovation

• Integrates causal attention for sequential recommendation
• Learns causal graph in user behavior sequences
• Refines attention mechanism using discovered causal graph