Title :
Regular Multiple Criteria Linear Programming for Semi-supervised Classification
Author :
Zhiquan Qi ; Yingjie Tian ; Yong Shi
Author_Institution :
Res. Center on Fictitious Econ. & Data Sci., Beijing, China
Abstract :
In this paper, inspired by the application potential of Regular Multiple Criteria Linear Programming (RMCLP), we proposed a novel Laplacian RMCLP(called Lap-RMCLP)method for semi-supervised classification problem, which can exploit the geometry information of the marginal distribution embedded in unlabeled data to construct a more reasonable classifier and is a useful extension of TSVM. Furthermore, by adjusting the parameter, Lap-RMCLP can convert to RMCLP naturally. All experiments on public and data sets and Basic Endowment Insurance Fund Audit(BEIFA) dataset show that Lap-RMCLP is a competitive method in semi-supervised classification.
Keywords :
geometry; insurance; learning (artificial intelligence); linear programming; pattern classification; support vector machines; BEIFA; TSVM; basic endowment insurance fund audit dataset; geometry information; marginal distribution; novel Laplacian RMCLP method; regular multiple criteria linear programming; semi-supervised classification problem; Accuracy; Kernel; Laplace equations; Linear programming; Manifolds; Support vector machines; Training; Basic Endowment Insurance Fund Audit (BEIFA) dataset; Laplacian; RMCLP; semi-supervised classification;
Conference_Titel :
Data Mining Workshops (ICDMW), 2012 IEEE 12th International Conference on
Conference_Location :
Brussels
Print_ISBN :
978-1-4673-5164-5
DOI :
10.1109/ICDMW.2012.65
