Proceedings of the Annual Conference of JSAI
Online ISSN : 2758-7347
32nd (2018)
Session ID : 4Pin1-41
Conference information

Why does the station get crowded irregularly in the near future?
*Tatsuo YAMASHITA, Kota TSUBOUCHI, Mikiya MARUYAMA, Yuuki YAMAURA, Koichiro OKADA
Author information
CONFERENCE PROCEEDINGS FREE ACCESS

Details
Abstract

In previous work of future congestion prediction using transit search logs, whether target station will get crowded on target time or not is predictable, but we cannot specify the cause of irregular congestion. In order to specify the irregular congestion cause, we analyzed posts containing information on future events from SNS. By combining congestion prediction information from the transit search log and event information from SNS, it was possible to extract the cause of congestion accurately.

Content from these authors
© 2018 The Japanese Society for Artificial Intelligence
Previous article Next article
feedback
Top