Abstract With the increase of the number of construction projects on soft ground due to the growth of regional industry, various construction problems related to soft soil behavior have also been reported. Especially, foundation piles of abutments and/or buildings which were constructed on soft ground have suffered from a great deal of stability problems due to excessive displacement accompanied by the lateral flow of soft ground. Although many researchers have studied this problem, it is still difficult to assess the mechanism of lateral flow quantitatively. Furthermore, a reliable design method for the judgment of lateral flow occurrence is not established yet. In this study, an advanced pattern recognition theory — i.e., the support vector machine (SVM) theory — was applied to develop a soft computing model for the judgment of lateral flow occurrence based on real data compiled from Korea and Japan. Predicted results from SVM models were compared with those from the conventional empirical methods. It has been found that the proposed SVM pattern recognition model can predict lateral flow occurrence practically and more accurately than the conventional empirical methods.
Development of advanced pattern recognition model for evaluation of lateral displacement on soft ground using support vector machine
KSCE Journal of Civil Engineering ; 14 , 2 ; 173-182
2010-03-01
10 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Support vector machine applied to the pattern recognition of activity of antagonists
British Library Online Contents | 2002
|Daily activity pattern recognition by using support vector machines with multiple classes
Online Contents | 2013
|Support vector machine applied to the pattern recognition of guanidines as Na/H exchange inhibitors
British Library Online Contents | 2002
|