Hossein Moosaei and Milan Hladík. Inverse free Universum twin support vector machine. In D. E. Simos and others, editors, Learning and Intelligent Optimization, LNCS, pp. 252–264, Springer, Cham, 2021.
[PDF] [gzipped postscript] [postscript] [HTML]
Universum twin support vector machine (U-TSVM) is an efficient method for binary classification problems. In this paper, we improve the U-TSVM algorithm and propose an improved Universum twin bounded support vector machine (named as IUTBSVM). Indeed, by introducing a different Lagrangian function for the primal problems, we obtain new dual formulations so that we do not need to compute inverse matrices. Also to reduce the computational time of the proposed method, we suggest smaller size of the rectangular kernel matrices than the other methods. Numerical experiments on several UCI benchmark data sets indicate that the IUTBSVM is more efficient than the other three algorithms, namely U-SVM, TSVM, and U-TSVM in sense of the classification accuracy.
@inCollection{MooHla2021b,
author = "Hossein Moosaei and Milan Hlad\'{\i}k",
title = "Inverse free Universum twin support vector machine",
editor = "D. E. Simos and others",
feditor = "Simos, Dimitris E. and Pardalos, Panos M. and Kotsireas, Ilias S.",
booktitle = "Learning and Intelligent Optimization",
fbooktitle = "Learning and Intelligent Optimization, 15th International Conference, LION 15, Athens, Greece, June 20-25, 2021, Revised Selected Papers",
publisher = "Springer",
address = "Cham",
series = "LNCS",
fseries = "Lecture Notes in Computer Science",
volume = "12931",
pages = "252-264",
year = "2021",
doi = "10.1007/978-3-030-92121-7_21",
isbn = "978-3-030-92121-7",
url = "https://doi.org/10.1007/978-3-030-92121-7_21",
bib2html_dl_html = "https://link.springer.com/chapter/10.1007/978-3-030-92121-7_21",
abstract = "Universum twin support vector machine (U-TSVM) is an efficient method for binary classification problems. In this paper, we improve the U-TSVM algorithm and propose an improved Universum twin bounded support vector machine (named as IUTBSVM). Indeed, by introducing a different Lagrangian function for the primal problems, we obtain new dual formulations so that we do not need to compute inverse matrices. Also to reduce the computational time of the proposed method, we suggest smaller size of the rectangular kernel matrices than the other methods. Numerical experiments on several UCI benchmark data sets indicate that the IUTBSVM is more efficient than the other three algorithms, namely U-SVM, TSVM, and U-TSVM in sense of the classification accuracy.",
keywords = "Support vector machine; Twin SVM; Universum data; U-SVM; U-TSVM",
}
Generated by bib2html.pl (written by Patrick Riley ) on Wed Oct 23, 2024 08:16:44