Using decision tree for data mining of pavement maintenance and management

Jyh Dong Lin, Wei Hsing Huang, Chia Tse Hung, Chien Ta Chen, Jih Chiang Lee

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

In recent years advancements in the Information Technology (IT), have enabled automation of pavement measurement data. A large amount of data can be saved for a pavement management system. The study of pavement maintenance and management has include many methods, such as" expert system, decision support analysis and data mining (DM) ". In this study we use decision tree for data mining algorithm C5.0 has been used in this analysis. After acceptance of the decision tree, we make use of algorithms and computing for classification. This method is used to check the pavement management system database and make a comparison of all data. The result shown a correct classification of about 61% it's still improved space. According to this result we discuss three analysis results included: 1.Database information is correct or not 2.Road pavement never homogenization 3.Milling process never remove human factor. Finally useful pavement information and ways can improve system integrity and correctly.

Original languageEnglish
Title of host publicationMaterials Engineering and Automatic Control II
Pages1015-1019
Number of pages5
DOIs
StatePublished - 2013
Event2nd International Conference on Materials Engineering and Automatic Control, ICMEAC 2013 - Shandong, China
Duration: 18 May 201319 May 2013

Publication series

NameApplied Mechanics and Materials
Volume330
ISSN (Print)1660-9336
ISSN (Electronic)1662-7482

Conference

Conference2nd International Conference on Materials Engineering and Automatic Control, ICMEAC 2013
Country/TerritoryChina
CityShandong
Period18/05/1319/05/13

Keywords

  • Data mining
  • Decision tree
  • Pavement maintenance and management

Fingerprint

Dive into the research topics of 'Using decision tree for data mining of pavement maintenance and management'. Together they form a unique fingerprint.

Cite this