Author : Prabhjot Kaur and Ravneet Kaur

As the information contained on the web is increasing from day to day, organizing this information could be a necessary requirement. Data mining process is to extract information from a data set and transform it into an understandable structure for further use. As each component in a Web page like HTML tags and terms is taken as a feature dimension of the classification problem becomes too high to be resolved by well-known classifiers decision trees and support vector machines etc. So we need efficient methods to select best features to reduce feature space of the Web page classification problem. In this study, a recent optimization technique namely the Membrane Computing (MC) is used, to select the best features. Membrane computing is an area within computer science, originate from natural computing. It is found that when features are selected by our membrane computing algorithm, J48 classifier is used to evaluate the fitness of selected features. WebKB datasets were classified without loss of accuracy. The experimental results of this study showed that, Membrane Computing algorithm is an acceptable optimization algorithm for Web Page feature selection.

Keywords: Membrane Computing, P Systems, Optimization, Classification, Feature selection, Web page classification.

Article published in International Journal of Current Engineering and Technology, Vol.4, No.5 (Oct-2014)