Title: Research on adaptive recommendation algorithm emerging in user interest in an electronic commerce environment

Authors: Qingqing Zhang; Xingjing Wu; Tinggui Chen

Addresses: College of Computer Science and Information Engineering, Zhejiang Gongshang University, Hangzhou, China ' College of Computer Science and Information Engineering, Zhejiang Gongshang University, Hangzhou, China ' College of Computer Science and Information Engineering, Zhejiang Gongshang University, Hangzhou, China

Abstract: The user's interest measuring, modelling and updating is the basic and core component in a personalised service system. In this paper, based on the factors that affect user's interest in combination, with the introduction of browsing bytes per unit time and improved calculation method of stay time on webpage, established a calculation function of interest degree with variables of browsing bytes per time, stay time with weight and browsing rate. According to the user's interest is stable in a short period of time, then set up the hierarchical vector space model (VSM) based on the representation of the user's interest model and the mechanism of updating. At the same time, interest degree was imported into vector model, and used the tree layers of tree structure to represent the user's interest. The interest subject is based on the characteristics of keywords, and the model was updated at a fixed interval time T, to improve the practicability of the personalised system, so as to better meet the user's personalised needs.

Keywords: interest degree calculation; user interest; modelling; vector space model; VSM; adaptive recommendation; electronic commerce; e-commerce; personalised services; personalisation; keywords; webpage stay time; browsing rate; recommender systems.

DOI: 10.1504/IJCSM.2015.072964

International Journal of Computing Science and Mathematics, 2015 Vol.6 No.5, pp.425 - 433

Received: 25 Mar 2015
Accepted: 06 May 2015
Published online: 10 Nov 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article