Understanding inter-region travel patterns is an important issue for many reasons. This paper aims at utilising the data of the number of travellers, which nowadays can be obtained easily at any time by using mobile phone location data. In this study, an automatic feature extraction method is proposed with a random matrix theory-based principal component analysis (RMT-PCA), and its ability is confirmed by applying it to the data of the number of long-period inter-region travellers. The results show that some seasonal and weekly patterns, as well as economic and climatic situations, were revealed by the data, some of which might be missed by a conventional method. In addition, the selection of data, whether daytime or nighttime, brought about a different result in both the number of extracted features and their interpretation.
Nakanishi, W., Yamaguchi, H., & Fukuda, D. (2018). Feature Extraction of Inter-Region Travel Pattern Using Random Matrix Theory and Mobile Phone Location Data. In Transportation Research Procedia (Vol. 34, pp. 115–122). Elsevier B.V. https://doi.org/10.1016/j.trpro.2018.11.022