Twin Support Vector Machines | springerprofessional.de

Springer Professional

nach oben

2017 | Buch

Kapitel lesen Erstes Kapitel lesen

Twin Support Vector Machines

Models, Extensions and Applications

verfasst von: Jayadeva, Reshma Khemchandani, Suresh Chandra

Verlag: Springer International Publishing

Buchreihe : Studies in Computational Intelligence

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

This book provides a systematic and focused study of the various aspects of twin support vector machines (TWSVM) and related developments for classification and regression. In addition to presenting most of the basic models of TWSVM and twin support vector regression (TWSVR) available in the literature, it also discusses the important and challenging applications of this new machine learning methodology. A chapter on “Additional Topics” has been included to discuss kernel optimization and support tensor machine topics, which are comparatively new but have great potential in applications. It is primarily written for graduate students and researchers in the area of machine learning and related topics in computer science, mathematics, electrical engineering, management science and finance.

Anzeige

Inhaltsverzeichnis

Frontmatter

Chapter 1. Introduction

Abstract

This chapter provides an overview of Support Vector Machines and some of its variants. We first discuss \(L_1\)-norm SVM and then proceed to discuss two of the most popular \(L_2\)-norm SVMs, namely the Least Squares SVM and the Proximal SVM. This chapter also discusses Support Vector Regression in the light of a significant contribution of Bi and Bennett that provides equivalence between a regression and appropriately constructed binary classification problem. Towards the end, we throw some light on certain efficient algorithms for solving SVMs and conclude with some limitations of SVMs.

Jayadeva, Reshma Khemchandani, Suresh Chandra

Chapter 2. Generalized Eigenvalue Proximal Support Vector Machines

Abstract

This chapter is devoted to the study of Generalized Eigenvalue Proximal Support Vector Machine (GEPSVM) and its variants. Both, classification as well as regression models are presented and their advantages over the traditional SVM and SVR formulations are discussed.

Jayadeva, Reshma Khemchandani, Suresh Chandra

Chapter 3. Twin Support Vector Machines (TWSVM) for Classification

Abstract

This chapter constitute the core around which the latter chapters are developed. Here the first and most basic formulation of the Twin Support Vector Machine (TWSVM) is presented. Certain advantages and some of the possible disadvantages/computational issues are noted. Two variants of TWSVM namely the Twin Bounded Support Vector Machine (TBSVM) and then Improved Twin Support Vector Machine (ITSVM), are discussed. These two variants attempt to handle some of the issues related with the original TWSVM formulation.

Jayadeva, Reshma Khemchandani, Suresh Chandra

Chapter 4. TWSVR: Twin Support Vector Machine Based Regression

Abstract

This chapter attempts to extend the twin methodology to the regression problem. Taking motivation from the celebrated Bi and Bennett’s result, the twin version of the regression formulation (TWSVR) is derived from TWSVM in exactly the same way as SVR is derived from SVM. Certain difficulties are observed in Peng’s TSVR formulation and it is shown that Peng’s formulation is not in the true spirit of twin methodology. This work is further extended to study the problems of approximating a function and its derivatives.

Jayadeva, Reshma Khemchandani, Suresh Chandra

Chapter 5. Variants of Twin Support Vector Machines: Some More Formulations

Abstract

Twin support vector machine formulation (TWSVM) is based on the idea of generalized eigenvalue proximal support vector machine formulation (GEPSVM), which determines two nonparallel planes by solving two related SVM-type problems, so that its computing cost in the training phase is much smaller than that of the standard SVM. In addition to keeping the superior characteristics of GEPSVM, the classification performance of TWSVM significantly outperforms that of GEPSVM. However, the stand-alone method requires the solution of two smaller quadratic programming problems. This chapter mainly reviews the research progress of TWSVM and presents some of its extensions including the learning model and specific applications in recent years.

Jayadeva, Reshma Khemchandani, Suresh Chandra

Chapter 6. TWSVM for Unsupervised and Semi-supervised Learning

Abstract

Semi-supervised learning falls between unsupervised learning (with no labeled information of training data) and supervised learning (with completely labeled training data). Recent researches in machine-learning has shown that unlabeled data, when used in conjunction with a small amount of labeled data, can produce considerable improvement in learning accuracy. This chapter discusses twin support vector machine based algorithms for unsupervised and semi-supervised framework.

Jayadeva, Reshma Khemchandani, Suresh Chandra

Chapter 7. Some Additional Topics

Abstract

This chapter is devoted to the study of certain additional topics associated with TWSVM. Specifically these are related with the problem of optimal kernel selection, knowledge based twin support vector machine and twin support tensor machines. The last topic, namely the twin support tensor machines, is very recent and has great potential in areas of text categorization and image processing.

Jayadeva, Reshma Khemchandani, Suresh Chandra

Chapter 8. Applications Based on TWSVM

Abstract

TWSVM and its variants have been widely discussed in earlier chapters. In this chapter, we tend to discuss the applications where the special properties of TWSVMs have been used. One of the major advantage of TWSVM is its superiority over other machine learning methodologies in dealing with unbalanced datasets. Such datasets naturally arise for e.g. in medical domain where samples of diseased patients is far less than normal patients resulting in unbalanced classes. The structure of TWSVM allows to identify the hyperplane close to samples of diseased patients. Thus via this chapter, we tend to make reader familiar with widespread applicability of TWSVM across a multitude of application domain.

Jayadeva, Reshma Khemchandani, Suresh Chandra

Backmatter

Titel: Twin Support Vector Machines
verfasst von: Jayadeva
Reshma Khemchandani
Suresh Chandra
Copyright-Jahr: 2017
Verlag: Springer International Publishing
Electronic ISBN: 978-3-319-46186-1
Print ISBN: 978-3-319-46184-7
DOI: https://doi.org/10.1007/978-3-319-46186-1

Weitere Formate
Fachgebiete
Bücher
Zeitschriften
Themenseiten
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
SLX-Digitalkonferenz: Zukunftswerkstatt 2024