A Self-Supervised Transformer for Unusable Shared Bike Detection

2025-05-02

AI Summary

To address label scarcity, class imbalance, and insufficient modeling capacity of static thresholds in fault detection for bike-sharing systems, this paper proposes a self-supervised Transformer framework tailored for spatiotemporal trajectory data. The method jointly leverages GPS trajectories and trip records to construct dynamic spatiotemporal feature representations, employing masked trajectory reconstruction for pretraining and lightweight binary classification for fine-tuning—enabling unsupervised anomaly discrimination. This work is the first to integrate self-supervised learning with Transformer encoders into bike-sharing fault identification, eliminating reliance on manual annotations and assumptions about prior distributions. Evaluated on real-world data from 10,730 bikes in Chengdu, the model achieves 97.81% accuracy, precision of 0.8889, and an F1-score of 0.9358—significantly outperforming conventional machine learning and supervised deep learning baselines.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Security and Privacy: Data transparency and provenance

Abstract

The rapid expansion of bike-sharing systems (BSS) has greatly improved urban"last-mile"connectivity, yet large-scale deployments face escalating operational challenges, particularly in detecting faulty bikes. Existing detection approaches either rely on static model-based thresholds that overlook dynamic spatiotemporal (ST) usage patterns or employ supervised learning methods that struggle with label scarcity and class imbalance. To address these limitations, this paper proposes a novel Self-Supervised Transformer (SSTransformer) framework for automatically detecting unusable shared bikes, leveraging ST features extracted from GPS trajectories and trip records. The model incorporates a self-supervised pre-training strategy to enhance its feature extraction capabilities, followed by fine-tuning for efficient status recognition. In the pre-training phase, the Transformer encoder learns generalized representations of bike movement via a self-supervised objective; in the fine-tuning phase, the encoder is adapted to a downstream binary classification task. Comprehensive experiments on a real-world dataset of 10,730 bikes (1,870 unusable, 8,860 normal) from Chengdu, China, demonstrate that SSTransformer significantly outperforms traditional machine learning, ensemble learning, and deep learning baselines, achieving the best accuracy (97.81%), precision (0.8889), and F1-score (0.9358). This work highlights the effectiveness of self-supervised Transformer on ST data for capturing complex anomalies in BSS, paving the way toward more reliable and scalable maintenance solutions for shared mobility.

Problem

• Detect faulty bikes in bike-sharing systems dynamically
• Overcome label scarcity and class imbalance issues
• Leverage spatiotemporal data for accurate anomaly detection

Innovation

• Self-Supervised Transformer for bike detection
• Leverages GPS trajectories and trip records
• Outperforms traditional ML and deep learning