Improved SVM Method Applied to the Online User Behavior Analysis

Author

Zuo, Lin

Author_Institution

Sch. of Energy Sci. & Eng., Univ. of Electron. Sci. & Technol. of China, Chengdu, China

fYear

2012

fDate

27-29 Oct. 2012

Firstpage

728

Lastpage

732

Abstract

Online user behavior analysis has gained extensive attention in recent years. In this paper, to obtain the real users´ online behaviors based on a DNS-level tracing approach, a new improved SVM (support vector machine) method for analyzing the users´ online behaviors is put forth, which enables to get insightful views at a large scale. As the increase of the amount of data, improving the convergence speed of SVM is highly desired. The computational efficiency of the proposed SVM of this work is greatly improved by rewriting KKT conditions for the Sequential Minimal Optimization (SMO) algorithm. The improved SVM possesses a great capability of clustering the users´ data and revealing the users´ behaviors accurately from various aspects. The effectiveness of the improved SVM method is validated and demonstrated via analyzing a set of data of users´ online behaviors.

Keywords

behavioural sciences; convergence; optimisation; pattern clustering; support vector machines; DNS-level tracing approach; KKT conditions; SMO algorithm; SVM convergence speed improving; computational efficiency; convergence speed improvement; online user behavior analysis; sequential minimal optimization algorithm; support vector machine method; user data clustering; Accuracy; Computational efficiency; Educational institutions; Electronic mail; Google; Optimization; Support vector machines; DNS; SMO; SVM; cluster; users¡¯ online behaviors;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer and Information Technology (CIT), 2012 IEEE 12th International Conference on

Conference_Location

Chengdu

Print_ISBN

978-1-4673-4873-7

Type

conf

DOI

10.1109/CIT.2012.150

Filename

6391987