Transferable Unsupervised Outlier Detection Framework for Human Semantic Trajectories

TL;DR

This paper introduces TOD4Traj, a domain-agnostic framework that leverages multi-modal data and contrastive learning to improve outlier detection in human semantic trajectories across diverse datasets.

Contribution

The paper proposes a novel transferable outlier detection framework that unifies multi-modal features and employs contrastive learning for better outlier identification.

Findings

Outperforms existing models in outlier detection accuracy

Effectively integrates multi-modal data for trajectory analysis

Demonstrates high adaptability across different datasets

Abstract

Semantic trajectories, which enrich spatial-temporal data with textual information such as trip purposes or location activities, are key for identifying outlier behaviors critical to healthcare, social security, and urban planning. Traditional outlier detection relies on heuristic rules, which requires domain knowledge and limits its ability to identify unseen outliers. Besides, there lacks a comprehensive approach that can jointly consider multi-modal data across spatial, temporal, and textual dimensions. Addressing the need for a domain-agnostic model, we propose the Transferable Outlier Detection for Human Semantic Trajectories (TOD4Traj) framework.TOD4Traj first introduces a modality feature unification module to align diverse data feature representations, enabling the integration of multi-modal information and enhancing transferability across different datasets. A contrastive learning module is further pro-posed for identifying regular mobility patterns both temporally and across populations, allowing for a joint detection of outliers based on individual consistency and group majority patterns. Our experimental results have shown TOD4Traj's superior performance over existing models, demonstrating its effectiveness and adaptability in detecting human trajectory outliers across various datasets.