A preliminary study on automatic algorithm selection for short-term traffic forecasting

Angarita-Zapata, Juan S.; Triguero, Isaac; Masegosa, Antonio D.

Abstract

© 2018, Springer Nature Switzerland AG. Despite the broad range of Machine Learning (ML) algorithms, there are no clear baselines to find the best method and its configuration given a Short-Term Traffic Forecasting (STTF) problem. In ML, this is known as the Model Selection Problem (MSP). Although Automatic Algorithm Selection (AAS) has proved success dealing with MSP in other areas, it has hardly been explored in STTF. This paper deepens into the benefits of AAS in this field. To this end, we have used Auto-WEKA, a well-known AAS method, and compared it to the general approach (which consists of selecting the best of a set of algorithms) over a multi-class imbalanced classification STTF problem. Experimental results show AAS as a promising methodology in this area and allow important conclusions to be drawn on how to improve the performance of ASS methods when dealing with STTF.

Citation

Angarita-Zapata, J. S., Triguero, I., & Masegosa, A. D. (2018, October). A preliminary study on automatic algorithm selection for short-term traffic forecasting. Presented at 12th International Symposium on Intelligent Distributed Computing (IDC 2018), Bilbao, Spain

• Presentation Conference Type: Edited Proceedings
• Conference Name: 12th International Symposium on Intelligent Distributed Computing (IDC 2018)
• Start Date: Oct 15, 2018
• End Date: Oct 17, 2018
• Acceptance Date: Jun 5, 2018
• Online Publication Date: Sep 15, 2018
• Publication Date: Jan 1, 2018
• Deposit Date: Oct 18, 2018
• Publicly Available Date: Jan 2, 2019
• Journal: Studies in Computational Intelligence
• Electronic ISSN: 1860-9503
• Publisher: Springer Publishing Company
• Peer Reviewed: Peer Reviewed
• Volume: 798
• Pages: 204-214
• Series Title: Studies in Computational Intelligence
• Series Number: 798
• Book Title: Intelligent Distributed Computing XII
• Chapter Number: 18
• ISBN: 9783319996257
• DOI: https://doi.org/10.1007/978-3-319-99626-4_18
• Public URL: https://nottingham-repository.worktribe.com/output/1174905
• Publisher URL: https://link.springer.com/chapter/10.1007/978-3-319-99626-4_18
• Contract Date: Oct 18, 2018