Abstract
The existing rail station passenger flow prediction models are inefficient, due to that most of them use single-source data to predict. In this paper, a novel method is proposed based on multi-layer LSTM, which integrates multi-source traffic data and multi-techniques (including feature selection based on Spearman correlation and time feature clustering), to improve the performance of predicting passenger flow. The experimental results show that the multi-source data and the techniques integrated in the model are helpful, and the proposed method obtains a higher prediction accuracy which outperforms other methods (e.g. SARIMA, SVR and BP network) greatly.
Highlights
With the rapid development of urban rail transit and the continuous improvement of information management system, a large number of passenger travel data have been generated
RELATED WORK There are many factors that affect the passenger flow of rail station, which could roughly be classified into two categories: temporal factors and spatial factors
3) SPATIAL FACTORS (1) Due to the connectivity of urban rail transit network, each rail station stw is definitely affected by its adjacent rail stations RailADSTw, i.e., PFlowRi−a1il,AjDSTw
Summary
With the rapid development of urban rail transit and the continuous improvement of information management system, a large number of passenger travel data have been generated. There are some problems in passenger flow forecasting currently, such as single data source, incomplete consideration of influencing factors, which yield low accuracy of existing methods, and seriously defect the management of the urban traffic. In this paper, based on multi-source traffic data, a novel method, which take both of temporal and spatial factors. Into consideration, is proposed to predict passenger flow of rail station It utilizes clustering algorithm in classifying features of temporal factor, due to that clustering algorithm, such as k-means, DBSCAN [1], [2], Density Peak [3], [4] etc., is an effective way to classify data into different categories automatically, and is quite suitable and applicable in this field. Due to the time series characteristics of passenger flow, based on the advantages of LSTM network [5] in modeling time series data, a multi-layer LSTM network passenger flow prediction model is proposed as well to predict the entrance passenger flow of rail station
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
