A Comparison of Machine Learning Techniques in the Carpooling Problem

• 作者: C. Méndez Santos E. Castán Rocha J. A. Castán Rocha J. D. Terán Villanueva J. Laria Menchaca J. Pérez Cobos M. A. Arteaga Santos M. G. Treviño Berrones S. Ibarra Martínez
• 来源期刊: 电脑和通信(英文)
• 学科: 工学
• 关键词: Carpooling Machine Learning Techniques Vehicle Traffic Congestion
• 年，卷（期）: 2020,（12）
• 页码范围: 159-169
• 页数: 11页
• 分类号: TN9

摘要

Urban traffic congestion is a severe and widely studied problem over the decade because of the negative impacts. However, in recent years some approaches emerge as proper and suitable solutions. The Carpooling initiative is one of the most representative efforts to propitiate a responsible use of particular vehicles. Thus, the paper introduces a carpooling model considering the users’ preference to reach an appropriate match among drivers and passengers. In particular, the paper conducts a study of 6 of the most avid classified techniques in machine learning to create a model for the selection of travel companions. The experimental results show the models’ precision and assess the best cases using Friedman’s test. Finally, the conclusions emphasize the relevance of the proposed study and suggest that it is necessary to extend the proposal with more drives and passengers’ data.