Authors: Chunhua Ju; Chonghuan Xu; Guanglan Zhou
Addresses: School of Management Science & e-Business, Zhejiang Gongshang University, Hangzhou, 310018, China ' School of Business Administration, Zhejiang Gongshang University, Hangzhou, 310018, China ' School of Statistics, Centre for Studies of Modern Business, Zhejiang Gongshang University, Hangzhou, 310018, China
Abstract: Owing to the convenience and accessibility of the online shopping, most adolescents are inclined to click the computer screen to order their daily necessities. Therefore, more and more e-business websites will provide personalised recommendation service on the basis of the user interest and browsing content. Customisation service gradually becomes a state-of-the-art research in the field of e-commerce user interest data mining. It tends to combine the user's individual context factors for the sake of tracking user interest trail, which helps classify the interest drift categories. According to the mentioned issues this article expounds the user context from context factor and user behaviour. Additionally, it initials the proposed method to explicate its multiple factor weights in order to determine the extent of user interest preference. Moreover, it establishes the hidden semi-Markov model to detect interest drift tracking via user viewing route. Finally, the proposed method is proved to be accurate and precise through the experimental analysis and verification.
Keywords: user behaviour; interest drift; individual context.
International Journal of Wireless and Mobile Computing, 2017 Vol.12 No.2, pp.117 - 123
Available online: 08 May 2017 *Full-text access for editors Access for subscribers Purchase this article Comment on this article