Development of traffic accidents prediction model with intelligent system theory

Soo Beom Lee, Tai Sik Lee, Hyung Jin Kim, Young Kyun Lee

Research output: Contribution to journalConference articlepeer-review

9 Scopus citations

Abstract

It is important to clarify the relationship between traffic accidents and various influencing factors in order to reduce the number of traffic accidents. This study developed a traffic accident frequency prediction model using multi-linear regression and quantification theories which are commonly applied in the field of traffic safety to verify the influences of various factors in the traffic accident frequency. The data was collected on the Korean National Highway 17 which shows the highest accident frequency and fatality in Chonbuk Province. In order to minimize the uncertainty of the data, the fuzzy theory and neural network theory were applied. The neural network theory can provide fair learning performance by modeling the human neural system mathematically. In conclusion, this study focused on the practicability of the fuzzy reasoning theory and the neural network theory for traffic safety analysis.

Original languageEnglish
Pages (from-to)880-888
Number of pages9
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3481
Issue numberII
DOIs
StatePublished - 2005
EventInternational Conference on Computational Science and Its Applications - ICCSA 2005 - , Singapore
Duration: 9 May 200512 May 2005

Fingerprint

Dive into the research topics of 'Development of traffic accidents prediction model with intelligent system theory'. Together they form a unique fingerprint.

Cite this