Traditional machine learning has been largely concerned with developing techniques for small or modestly sized datasets. These techniques fail to scale up well for large data, a situation becoming increasingly common in today’s world. Furthermore most of the machine learning classifiers are trained in a batch way. Under this model, all training data is given a priori and training is performed in one batch. If more training data is later obtained the classifier must be re-trained from scratch. Re-solving the problem from scratch seems computationally wasteful. In this research we will focus on developing classifier for big data sets and incremental way of learning for dealing with real world problem. In this paper we propose a feature extraction and classification algorithm for big data which is based on incremental kernel PCA and conjugate gradient based LS-SVM. Through experimental results on big data in UCI machine learning repository we can show that proposed classification algorithm enables solving large scale classification problems.

Unable to display preview. Download preview PDF.

Authors and Affiliations

1. Department of Computer Engineering, Youngsan University, Korea

   Byung Joo Kim

Editors and Affiliations

1. Department of Computer Engineering, Hannam University, 70 Hannamro, Daedeuk-gu, Daejeon, Korea

   Geuk Lee

2. Computer Science and Information Systems, University of Limerick, Limerick, Ireland

   Daniel Howard

3. University of Warsaw and Infobright Inc., Poland

   Dominik Ślęzak

4. School of Information and Communication Engineering, Sang JI University, Wonju, Kangwon, Korea

   You Sik Hong

Reprints and permissions

© 2012 Springer-Verlag Berlin Heidelberg

Policies and ethics