RT info:eu-repo/semantics/conferenceObject
T1 A prediction of bike flow in bike renting systems with the tensor model and deep learning
A1 Seitbekova, Yerkezhan
A1 Assilbekov, Bakytzhan
A1 Kuljabekov, Alibek
A1 Beisembetov, Iskander
K1 Bicicletas
K1 Bicycles
K1 Ingeniería civil
K1 Civil engineering
K1 Transportes
K1 Transportation
AB Rental bikes are popular in many urban areas to help people expand their mobility. It isimportant to make the rental bicycle usable and available to the general public at theappropriate time and place. Inevitably, providing the city with a steady supply of rentalbicycles becomes a major concern. The most important aspect is the estimation of thenumber of bicycles required in each bicycle sharing station at any given hour. This papergives an examination of human mobility as indicated by bicycle renting information of thebike sharing system. In this paper, we proposed a new approach for forecasting the bikeinflow and outflow from one station to another during certain time slots. Our methodanalyses human mobility pattern by two steps: (1) Using Tuckers tensor decomposition tocreate a 3D tensor to model human mobility and extract latent temporal and spatialcharacteristics of various stations and time slots. (2) to use a Long-Short Term MemoryNeural Network to model the relationship between mobility patterns and the derived latentspatial and temporal features in order to predict bike flow between stations. The maincontribution of this study that with the extracted latent characteristics through Tuckersfactorization we improve the accuracy of prediction by 16% and decrease the amount oftraining data that used in prediction. Also, a root mean squared error of prediction is 1,5bike.We compare our model with baseline models as historical average, ARMA, the feed-forwardneural network, and KNN. The proposed method showed the best results.
PB Universidad de Burgos. Servicio de Publicaciones e Imagen Institucional
SN 978-84-18465-12-3
YR 2021
FD 2021-07
LK http://hdl.handle.net/10259/6858
UL http://hdl.handle.net/10259/6858
LA eng
NO Trabajo presentado en: R-Evolucionando el transporte, XIV Congreso de Ingeniería del Transporte (CIT 2021), realizado en modalidad online los días 6, 7 y 8 de julio de 2021, organizado por la Universidad de Burgos
NO This work has done under the project № АР05134776«Location Analytics Techniques for Prediction of Mobility Patterns» of the Ministry of Education and Sciences of the Republic of Kazakhstan.
DS Repositorio Institucional de la Universidad de Burgos
RD 27-nov-2024