TY - GEN
T1 - Determination of collection points of bus probe data to achieve high prediction performance and low collection cost
AU - Kawatani, Takuya
AU - Mine, Tsunenori
N1 - Funding Information:
ACKNOWLEDGMENT This research was partially supported by JSPS Grant-in-Aid for Scientific Research No. JP19KK0257 and JP21H00907. We would like to thank Showa Bus Co., Ltd. for their cooperation and permission to analyze the data. We would like to express our gratitude to them.
Publisher Copyright:
© 2021 IPSJ.
PY - 2021
Y1 - 2021
N2 - The analysis and understanding of the operation status of bus routes are very effective in improving bus transportation services. However, the financial cost of obtaining real-time or near-real-time operation status for small and medium bus companies cannot be ignored. Here, ETC 2.0-based system can reduce the financial cost of system construction and operation. In this paper, we address the problem of determining the location of a roadside device to obtain the highest performance of travel time prediction; we conducted extensive experiments on a real dataset, namely city bus probe data collected over one year from April 2019 to March 2020. The experiments show that the location point with the largest travel time variance does not always determine the best prediction performance.
AB - The analysis and understanding of the operation status of bus routes are very effective in improving bus transportation services. However, the financial cost of obtaining real-time or near-real-time operation status for small and medium bus companies cannot be ignored. Here, ETC 2.0-based system can reduce the financial cost of system construction and operation. In this paper, we address the problem of determining the location of a roadside device to obtain the highest performance of travel time prediction; we conducted extensive experiments on a real dataset, namely city bus probe data collected over one year from April 2019 to March 2020. The experiments show that the location point with the largest travel time variance does not always determine the best prediction performance.
UR - http://www.scopus.com/inward/record.url?scp=85123908623&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85123908623&partnerID=8YFLogxK
U2 - 10.23919/ICMU50196.2021.9638959
DO - 10.23919/ICMU50196.2021.9638959
M3 - Conference contribution
AN - SCOPUS:85123908623
T3 - 13th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2021
BT - 13th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 13th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2021
Y2 - 17 November 2021 through 19 November 2021
ER -
