User Re-Identification via Confusion of the Contrastive Distillation Network and Attention Mechanism
Author:
Zhang Mingming1, Wang Bin1, Zhu Sulei1, Zhou Xiaoping1, Yang Tao2, Zhai Xi2
Affiliation:
1. College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 200234, China 2. Shanghai Urban and Rural Construction and Traffic Development Research Institute, Shanghai 200234, China
Abstract
With the rise of social networks, more and more users share their location on social networks. This gives us a new perspective on the study of user movement patterns. In this paper, we solve the trajectory re-identification task by identifying human movement patterns and then linking unknown trajectories to the user who generated them. Existing solutions generally focus on the location point and the location point information, or a single trajectory, and few studies pay attention to the information between the trajectory and the trajectory. For this reason, in this paper, we propose a new model based on a contrastive distillation network, which uses a contrastive distillation model and attention mechanisms to capture latent semantic information for trajectory sequences and focuses on common key information between pairs of trajectories. Combined with the trajectory library composed of historical trajectories, it not only reduces the number of candidate trajectories but also improves the accuracy of trajectory re-identification. Our extensive experiments on three real-world location-based social network (LBSN) datasets show that our method outperforms existing methods.
Subject
Electrical and Electronic Engineering,Biochemistry,Instrumentation,Atomic and Molecular Physics, and Optics,Analytical Chemistry
Reference38 articles.
1. Big data and its applications in urban intelligent transportation system;Lu;J. Transp. Syst. Eng. Inf. Technol.,2015 2. Dai, S., Wang, J., Huang, C., Yu, Y., and Dong, J. (2021, January 7–10). Temporal multi-view graph convolutional networks for citywide traffic volume inference. Proceedings of the 2021 IEEE International Conference on Data Mining (ICDM), Auckland, New Zealand. 3. Ye, M., Yin, P., Lee, W.-C., and Lee, D.-L. (2011, January 11–15). Exploiting geographical influence for collaborative point-of-interest recommendation. Proceedings of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval, Online. 4. Lim, K.H., Chan, J., Leckie, C., and Karunasekera, S. (2015, January 25–31). Personalized tour recommendation based on user interests and points of interest visit durations. Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, Buenos Aires, Argentina. 5. Hu, B., and Ester, M. (2014, January 14–17). Social topic modeling for point-of-interest recommendation in location-based social networks. Proceedings of the 2014 IEEE International Conference on Data Mining, Shenzhen, China.
|
|