Title: Identifying web navigation behaviour and patterns automatically from clickstream data

Authors: I-Hsien Ting, Lillian Clark, Chris Kimble

Addresses: Department of Information Management, National University of Kaohsiung, No. 700 Kaohsiung University Road, 811, Kaohsiung City, Taiwan. ' Department of Human Resource and Marketing Management, Portsmouth Business School, University of Portsmouth, Richmond Building, Portland Street, Portsmouth PO1 3DE, UK. ' Management Information Systems, Euromed Marseille Ecole de Management, Domaine de Luminy, BP 921, 13288, Marseille Cedex 9, France

Abstract: A user|s clickstream, such as that which is found in server-side logs, can be a rich source of data concerning the ways in which a user navigates a site, but the volume and level of detail found in these logs makes it difficult to identify and categorise specific navigational patterns. In this paper, we describe the three-step automatic pattern discovery (APD) method, a tool that utilises sequential mining to extract a user|s navigation route based on two levels of basic navigational elements. This paper contains descriptions of two studies in which the APD was used; the first makes use of APD to analyse the usage of an educational website; the second describes how APD was used to improve the design of a technical support website in a university department.

Keywords: navigation behaviour; web usage mining; clickstream data; sequential mining; web navigation; navigation patterns; automatic pattern discovery; educational website; technical support website.

DOI: 10.1504/IJWET.2009.032255

International Journal of Web Engineering and Technology, 2009 Vol.5 No.4, pp.398 - 426

Published online: 18 Mar 2010 *

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