DiffMove: Group Mobility Tendency Enhanced Trajectory Recovery via Diffusion Model

2025-03-24

AI Summary

Real-world trajectory data are often highly sparse due to low sampling rates or incomplete spatial coverage, posing two key challenges for trajectory recovery: insufficient historical information and difficulty in modeling individual mobility preferences. To address these, we propose a conditional diffusion model that jointly incorporates population-level trends and individual-specific preferences. Specifically, we first construct a graph neural network–driven population tendency graph; then, we jointly encode spatiotemporal characteristics via positional embeddings and multi-view individual preference representations, integrating both into the diffusion process. Our method is the first to enable dynamic coupling of population priors and fine-grained individual preferences during denoising, overcoming the limitations of weak supervision and ambiguous preference modeling in sparse regimes. Evaluated on two real-world datasets, our approach achieves significant improvements in localization accuracy and semantic coherence over state-of-the-art methods, while demonstrating strong robustness across varying sparsity levels.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Knowledge Representation and Reasoning: Preferences
• Search and Optimization: Sampling/Simulation-based Search

Application Category

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

Abstract

In the real world, trajectory data is often sparse and incomplete due to low collection frequencies or limited device coverage. Trajectory recovery aims to recover these missing trajectory points, making the trajectories denser and more complete. However, this task faces two key challenges: 1) The excessive sparsity of individual trajectories makes it difficult to effectively leverage historical information for recovery; 2) Sparse trajectories make it harder to capture complex individual mobility preferences. To address these challenges, we propose a novel method called DiffMove. Firstly, we harness crowd wisdom for trajectory recovery. Specifically, we construct a group tendency graph using the collective trajectories of all users and then integrate the group mobility trends into the location representations via graph embedding. This solves the challenge of sparse trajectories being unable to rely on individual historical trajectories for recovery. Secondly, we capture individual mobility preferences from both historical and current perspectives. Finally, we integrate group mobility tendencies and individual preferences into the spatiotemporal distribution of the trajectory to recover high-quality trajectories. Extensive experiments on two real-world datasets demonstrate that DiffMove outperforms existing state-of-the-art methods. Further analysis validates the robustness of our method.

Problem

• Recover sparse incomplete trajectory data effectively
• Leverage group mobility trends for trajectory recovery
• Capture individual mobility preferences accurately

Innovation

• Uses group tendency graph for trajectory recovery
• Integrates group and individual mobility preferences
• Employs diffusion model for high-quality recovery