CausalMob: Causal Human Mobility Prediction with LLMs-derived Human Intentions toward Public Events

2024-12-03

AI Summary

Rare public events (e.g., disasters, celebrations) causally disrupt human mobility patterns, yet existing models fail to explicitly model such causal interventions, leading to biased predictions under event-driven confounding. Method: We propose a causal-enhanced prediction framework that (i) leverages large language models (LLMs) to extract and formalize human intent from news texts as interpretable causal intervention variables; (ii) designs an orthogonalized causal inference architecture—decoupling event features from confounders via multi-source spatiotemporal graph neural networks and causal representation learning; and (iii) employs doubly robust estimation for deconfounded causal effect estimation. Contribution/Results: Evaluated on large-scale real-world datasets, our model reduces average prediction error by 18.7% overall and achieves up to 32.4% improvement in event-sensitive scenarios—significantly outperforming state-of-the-art methods. This validates the efficacy and generalizability of intent-driven causal modeling for human mobility forecasting.

Technology Category

• Machine Learning: Causal Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Reasoning under Uncertainty: Causality

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• 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

Abstract

Large-scale human mobility exhibits spatial and temporal patterns that can assist policymakers in decision making. Although traditional prediction models attempt to capture these patterns, they often interfered by non-periodic public events, such as disasters and occasional celebrations. Since regular human mobility patterns are heavily affected by these events, estimating their causal effects is critical to accurate mobility predictions. Although news articles provide unique perspectives on these events in an unstructured format, processing is a challenge. In this study, we propose a causality-augmented prediction model, called CausalMob, to analyze the causal effects of public events. We first utilize large language models (LLMs) to extract human intentions from news articles and transform them into features that act as causal treatments. Next, the model learns representations of spatio-temporal regional covariates from multiple data sources to serve as confounders for causal inference. Finally, we present a causal effect estimation framework to ensure event features remain independent of confounders during prediction. Based on large-scale real-world data, the experimental results show that the proposed model excels in human mobility prediction, outperforming state-of-the-art models.

Problem

• Predictive modeling
• Rare events
• Activity patterns

Innovation

• CausalMob model
• text analysis
• predictive performance