Laplacian Twin Support Vector Machine With Pinball Loss for Semi-Supervised Classification
Open Access
- 27 March 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 31399-31416
Abstract
Semi-supervised learning utilizes labeled data and the geometric information in the unlabeled data to construct a model whereas supervised learning makes use of the only label data. So, semi-supervised learning establishes a more reasonable classifier. In recent years, the Laplacian support vector machine (Lap-SVM) has received a lot of interest in the framework of semi-supervised classification. To develop the performance of Lap-SVM, Laplacian twin support vector machine (Lap-TSVM) has shown exceptional performance as an addition to improve the computational complexity. However, dealing with noise sensitivity and instability for resampling due to its hinge loss function is still a challenge. In this paper, we provide a novel Laplacian twin support vector machine by combining pinball loss function, termed as Lap-PTSVM, to effectively handle the aforementioned problems. As a result, it improves a better generalization ability of the classifier. Several experiments have been performed on artificial and UCI datasets. The results show that the proposed model has noise insensitivity comparable to the Lap-TSVM and has a great generalization performance. Furthermore, non-parametric statistical test are also conducted to justify the competitive performance of Lap-PTSVM.Keywords
Funding Information
- National Science, Research and Innovation Fund (NSRF) via the Program Management Unit for Human Resources and Institutional Development, Research and Innovation, Thailand (B05F640180)
This publication has 35 references indexed in Scilit:
- Laplacian least squares twin support vector machine for semi-supervised classificationNeurocomputing, 2014
- Support Vector Machine Classifier With Pinball LossIEEE Transactions on Pattern Analysis and Machine Intelligence, 2013
- Laplacian twin support vector machine for semi-supervised classificationNeural Networks, 2012
- On the effectiveness of preprocessing methods when dealing with different levels of class imbalanceKnowledge-Based Systems, 2012
- Feature selection in the Laplacian support vector machineComputational Statistics & Data Analysis, 2011
- Advanced nonparametric tests for multiple comparisons in the design of experiments in computational intelligence and data mining: Experimental analysis of powerInformation Sciences, 2010
- Optimal kernel selection in twin support vector machinesOptimization Letters, 2008
- An overview of statistical learning theoryIEEE Transactions on Neural Networks, 1999
- Support-vector networksMachine Learning, 1995
- Solutions of Ill-Posed Problems (A. N. Tikhonov and V. Y. Arsenin)SIAM Review, 1979