Deep learning vs. traditional solutions for group trajectory outliers

Belhadi, Asma; Djenouri, Youcef; Djenouri, Djamel; Michalak, Tomasz; Chun-Wei Lin, Jerry

Abstract

This article introduces a new model to identify a group of trajectory outliers from a large trajectory database and proposes several algorithms. These can be split into three categories: 1) algorithms based on data mining and knowledge discovery, which study the different correlations among the trajectory data and identify the group of abnormal trajectories from the knowledge extracted; 2) algorithms based on machine learning and computational intelligence methods, which use the ensemble learning and metaheuristics to find the group of trajectory outliers; and 3) an algorithm exploring the convolution deep neural network that learns the different features of historical data to determine the group of trajectory outliers. Experiments on different trajectory databases have been carried out to investigate the proposed algorithms. The results show that the deep learning solution outperforms data mining, machine learning, and computational intelligence solutions, as well as state-of-the-art solutions in terms of runtime and accuracy performance.

Citation

Belhadi, A., Djenouri, Y., Djenouri, D., Michalak, T., & Chun-Wei Lin, J. (2022). Deep learning vs. traditional solutions for group trajectory outliers. IEEE Transactions on Cybernetics, 52(6), 4508-4519. https://doi.org/10.1109/TCYB.2020.3029338

• Journal Article Type: Article
• Acceptance Date: Feb 1, 2021
• Online Publication Date: Nov 17, 2020
• Publication Date: 2022-06
• Deposit Date: Apr 22, 2021
• Publicly Available Date: Apr 23, 2021
• Journal: IEEE Transactions on Cybernetics
• Print ISSN: 2168-2267
• Electronic ISSN: 2168-2275
• Publisher: Institute of Electrical and Electronics Engineers
• Peer Reviewed: Peer Reviewed
• Volume: 52
• Issue: 6
• Pages: 4508-4519
• DOI: https://doi.org/10.1109/TCYB.2020.3029338
• Keywords: Computational intelligence; data mining; deep learning; machine learning; trajectory data
• Public URL: https://uwe-repository.worktribe.com/output/7283564