DSpace at Cochin University >
Department of Computer Science >
Seminar Reports >
MTech 2009-2011 Batch >
Please use this identifier to cite or link to this item:
|Title: ||Training Algorithms for Support Vector Machines|
|Authors: ||MRIDULA, T V|
|Keywords: ||Support Vector Machine|
Minimal optimization (SMO)
Statistical Learning Theory
|Issue Date: ||14-Jun-2010|
|Abstract: ||The support vector machine (SVM) is a training algorithm for learning classification and regression rules from data, for example the SVM can be used to learn polynomial and radial basis function (RBF). SVMs were first suggested by Vapnik in the 1990s for classification and have recently become an area of intense research owing to developments in the techniques and theory coupled with extensions to regression and classification.
SVMs arose from statistical learning theory; the aim being to solve only the problem of interest without solving a more difficult problem as an intermediate step. Two key elements in the implementation of SVM are the techniques of mathematical programming and kernel functions. The parameters are found by solving a quadratic programming problem with linear equality and inequality constraints; rather than by solving a non-convex, unconstrained optimization problem. The flexibility of kernel functions allows the SVM to search a wide variety of hypothesis spaces.
This seminar focuses on SVMs for two-class classification, The geometrical interpretation of support vector classification (SVC) is that the algorithm searches for the optimal separating surface, i.e. the hyperplane that is, in a sense, equidistant from the two classes. This optimal separating hyperplane has many nice statistical properties. SVC is outlined first for the linearly separable case. Kernel functions are then introduced in order to construct non-linear decision surfaces. Finally, for noisy data, when complete separation of the two classes may not be desirable, slack variables are introduced to allow for training errors.|
|Appears in Collections:||MTech 2009-2011 Batch|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.