Finding A Taxi with Illegal Driver Substitution Activity via Behavior Modelings
Junbiao Pang, Muhammad Ayub Sabir, Zhuyun Wang, Anjing Hu, Xue Yang, Haitao Yu, Qingming Huang
TL;DR
This work addresses the problem of identifying taxis involved in Illegal Driver Substitution (IDS) by formulating IDS detection as supervised learning over GPS traces and taximeter records. It introduces two driver-behavior models—Sleeping Time and Location (STL) and Pick-Up (PU)—and maps their outputs into a common discriminative feature space using Self-Similarity and Multi-Scale Pooling (MSP-SS). A novel MC-MIL framework fuses STL- and PU-based classifiers to align IDS-related patterns across long time horizons, showing superior performance over MIL-based baselines on the IDS@BJ dataset. The proposed approach yields robust, generalizable detection with practical efficiency, enabling targeted law-enforcement inspections in real-world settings. The work demonstrates meaningful improvements over baselines and provides actionable insights for deploying GPS–taximeter data fusion in transportation safety applications.
Abstract
In our urban life, Illegal Driver Substitution (IDS) activity for a taxi is a grave unlawful activity in the taxi industry, possibly causing severe traffic accidents and painful social repercussions. Currently, the IDS activity is manually supervised by law enforcers, i.e., law enforcers empirically choose a taxi and inspect it. The pressing problem of this scheme is the dilemma between the limited number of law-enforcers and the large volume of taxis. In this paper, motivated by this problem, we propose a computational method that helps law enforcers efficiently find the taxis which tend to have the IDS activity. Firstly, our method converts the identification of the IDS activity to a supervised learning task. Secondly, two kinds of taxi driver behaviors, i.e., the Sleeping Time and Location (STL) behavior and the Pick-Up (PU) behavior are proposed. Thirdly, the multiple scale pooling on self-similarity is proposed to encode the individual behaviors into the universal features for all taxis. Finally, a Multiple Component- Multiple Instance Learning (MC-MIL) method is proposed to handle the deficiency of the behavior features and to align the behavior features simultaneously. Extensive experiments on a real-world data set shows that the proposed behavior features have a good generalization ability across different classifiers, and the proposed MC-MIL method suppresses the baseline methods.