Causality-Aware Next Location Prediction Framework based on Human Mobility Stratification

2025-03-23

AI Summary

Existing next-location prediction methods overlook causal structures inherent in human mobility data, rendering them susceptible to confounding biases. To address this, we propose the first causally-aware framework integrating mobility hierarchy with causal graph learning. Our approach first constructs an interpretable causal graph to explicitly model causal relationships among users, locations, and time. Second, it incorporates counterfactual reasoning to explicitly disentangle confounding effects, thereby improving modeling of indirect influences—particularly for non-anchor-directed trips. Third, it features a plug-and-play module enabling seamless integration of diverse base models. Evaluated on multiple real-world datasets, our framework consistently improves Recall@10 by 6.2% on average across mainstream models. Ablation studies confirm both the efficacy and enhanced interpretability conferred by the learned causal graph.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Human mobility data are fused with multiple travel patterns and hidden spatiotemporal patterns are extracted by integrating user, location, and time information to improve next location prediction accuracy. In existing next location prediction methods, different causal relationships that result from patterns in human mobility data are ignored, which leads to confounding information that can have a negative effect on predictions. Therefore, this study introduces a causality-aware framework for next location prediction, focusing on human mobility stratification for travel patterns. In our research, a novel causal graph is developed that describes the relationships between various input variables. We use counterfactuals to enhance the indirect effects in our causal graph for specific travel patterns: non-anchor targeted travels. The proposed framework is designed as a plug-and-play module that integrates multiple next location prediction paradigms. We tested our proposed framework using several state-of-the-art models and human mobility datasets, and the results reveal that the proposed module improves the prediction performance. In addition, we provide results from the ablation study and quantitative study to demonstrate the soundness of our causal graph and its ability to further enhance the interpretability of the current next location prediction models.

Problem

• Improves next location prediction using causality-aware framework
• Addresses ignored causal relationships in human mobility data
• Enhances interpretability of prediction models via causal graphs

Innovation

• Integrates user, location, time for spatiotemporal patterns
• Develops causality-aware framework with novel causal graph
• Uses counterfactuals to enhance indirect travel effects