Development of neural network based traffic flow predictors using pre-processed data

Neural networks have commonly been applied for traffic flow predictions. Generally, the past traffic flow data captured by on-road detector stations, is used to train the neural networks. However, recently research mostly focuses on development of innovative neural networks, while it lacks developme...

Full description

Bibliographic Details
Main Authors: Chan, Kit Yan, Yiu, Ka Fai
Other Authors: Honglei Xu
Format: Book Chapter
Published: Springer 2014
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/18134
_version_ 1848749657909362688
author Chan, Kit Yan
Yiu, Ka Fai
author2 Honglei Xu
author_facet Honglei Xu
Chan, Kit Yan
Yiu, Ka Fai
author_sort Chan, Kit Yan
building Curtin Institutional Repository
collection Online Access
description Neural networks have commonly been applied for traffic flow predictions. Generally, the past traffic flow data captured by on-road detector stations, is used to train the neural networks. However, recently research mostly focuses on development of innovative neural networks, while it lacks development of mechanisms on pre-processing traffic flow data priors on training in order to obtain more accurate neural networks. In this chapter, a simple but effective training method is proposed by incorporating the mechanisms of back-propagation algorithm and the exponential smoothing method, which is proposed to pre-process traffic flow data before training purposes. The pre-processing approach intends to aid the back-propagation algorithm to develop more accurate neural networks, as the pre-processed traffic flow data is more smooth and continuous than the original unprocessed traffic flow data. This approach was evaluated based on some sets of traffic flow data captured on a section of the freeway in Western Australia. Experimental results indicate that the neural networks developed based on this pre-processed data outperform those that are developed based on either original data or data which is preprocessed by the other pre-processing approaches.
first_indexed 2025-11-14T07:24:26Z
format Book Chapter
id curtin-20.500.11937-18134
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T07:24:26Z
publishDate 2014
publisher Springer
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-181342023-02-13T08:01:38Z Development of neural network based traffic flow predictors using pre-processed data Chan, Kit Yan Yiu, Ka Fai Honglei Xu Xiangyu Wang Data cleansing Data pre-processing Time-series forecasting Traffic flow predictions Neural network Intelligent traffic management Neural networks have commonly been applied for traffic flow predictions. Generally, the past traffic flow data captured by on-road detector stations, is used to train the neural networks. However, recently research mostly focuses on development of innovative neural networks, while it lacks development of mechanisms on pre-processing traffic flow data priors on training in order to obtain more accurate neural networks. In this chapter, a simple but effective training method is proposed by incorporating the mechanisms of back-propagation algorithm and the exponential smoothing method, which is proposed to pre-process traffic flow data before training purposes. The pre-processing approach intends to aid the back-propagation algorithm to develop more accurate neural networks, as the pre-processed traffic flow data is more smooth and continuous than the original unprocessed traffic flow data. This approach was evaluated based on some sets of traffic flow data captured on a section of the freeway in Western Australia. Experimental results indicate that the neural networks developed based on this pre-processed data outperform those that are developed based on either original data or data which is preprocessed by the other pre-processing approaches. 2014 Book Chapter http://hdl.handle.net/20.500.11937/18134 10.1007/978-94-017-8044-5_8 Springer restricted
spellingShingle Data cleansing
Data pre-processing
Time-series forecasting
Traffic flow predictions
Neural network
Intelligent traffic management
Chan, Kit Yan
Yiu, Ka Fai
Development of neural network based traffic flow predictors using pre-processed data
title Development of neural network based traffic flow predictors using pre-processed data
title_full Development of neural network based traffic flow predictors using pre-processed data
title_fullStr Development of neural network based traffic flow predictors using pre-processed data
title_full_unstemmed Development of neural network based traffic flow predictors using pre-processed data
title_short Development of neural network based traffic flow predictors using pre-processed data
title_sort development of neural network based traffic flow predictors using pre-processed data
topic Data cleansing
Data pre-processing
Time-series forecasting
Traffic flow predictions
Neural network
Intelligent traffic management
url http://hdl.handle.net/20.500.11937/18134