On the study of GRBF and polynomial kernel based support vector machine in web logs

Author

Sahoo, P. ; Behera, Ajit Kumar ; Pandia, Manoj Kumar ; Dash, C. Sanjeev Kumar ; Dehuri, S.

Author_Institution

Dept. of Comput. Sci., Silicon Inst. of Technol., Bhubaneswar, India

fYear

2013

fDate

13-14 Sept. 2013

Firstpage

1

Lastpage

5

Abstract

World Wide Web continues to grow day-by-day and it is difficult to track and understand users´ need for the owners of a website. Therefore, an intelligent analyzer is required to find out the browsing patterns of a user. Moreover, the pattern which is revealed from this surge of web access logs must be useful, motivating, and logical. In this paper, two different kernel functions of support vector machine (SVM) are used to classify the web pages based on access time and region. Additionally, kernel parameters are also varied to study the trends of the accuracy of classification. Experimental results reveal that Gaussian radial basis function (GRBF) kernel based S VM is performing better than the polynomial kernel based SVM.

Keywords

Internet; data mining; document handling; pattern classification; radial basis function networks; support vector machines; GRBF kernel based support vector machine; Gaussian radial basis function; SVM; Web document repository; Web logs; Web mining; Web page classification; Web site; World Wide Web; access time; browsing patterns; intelligent analyzer; kernel parameters; polynomial kernel based support vector machine; region; Accuracy; Data mining; Kernel; Polynomials; Support vector machines; Training; Web pages; Classification; Radial basis functions; Support vector machine; Web log; World wide web;

fLanguage

English

Publisher

ieee

Conference_Titel

Emerging Trends and Applications in Computer Science (ICETACS), 2013 1st International Conference on

Conference_Location

Shillong

Print_ISBN

978-1-4673-5249-9

Type

conf

DOI

10.1109/ICETACS.2013.6691384

Filename

6691384