Case Study on the Determination of Building Materials Using a Support Vector Machine

Title
Case Study on the Determination of Building Materials Using a Support Vector Machine
Author(s)
김상용김중섭[김중섭]Llewellyn Tang[Llewellyn Tang]
Keywords
NEURAL-NETWORKS; SELECTION
Issue Date
201403
Publisher
ASCE-AMER SOC CIVIL ENGINEERS
Citation
JOURNAL OF COMPUTING IN CIVIL ENGINEERING, v.28, no.2, pp.315 - 326
Abstract
For any construction project to succeed, it is very important to select the materials accurately during the project's initial stage. Trying to choose the best-performing materials is a crucial task for the successful completion of a construction project. The material selection process typically is performed through the information received from a highly experienced decision maker and a purchasing agent without the logical decision making; thus, the construction field gains access to various artificial intelligence (AI) techniques to support decision models in their own selection method. Through a case study, this paper proposes the application of a systematic and efficient support vector machine (SVM) model to select suitable materials. The 120 data sets of the case study have completed building projects in South Korea. These data set were divided into three groups and constructed five binary classification models in the one-against-all (OAA) multiclassification method by data classification and normalization, resulting in the SVM model, based on the kernel polynominal, yielding a prediction accuracy rate of 87.5%. This case study indicates that the SVM model appears feasible to be the decision support model for selecting construction methods.
URI
http://hdl.handle.net/YU.REPOSITORY/32840http://dx.doi.org/10.1061/(ASCE)CP.1943-5487.0000259
ISSN
0887-3801
Appears in Collections:
건축학부 > 건축학부 > Articles
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE