Identifying and Using Hypermedia Browsing Patterns

Hypermedia offers benefits for users who wish to find information and users who wish to learn about a particular topic (Jonassen & Grabinger, 1991). However, hypermedia is also plagued by drawbacks that were identified early in its inception (Conklin, 1987) but which are still present (Theng & Thimbleby, 1998). Recent hypermedia research has focused on alleviating one of the classic problems of hypermedia, the problem of the user becoming lost while using it. One such approach is called Adaptive-Hypermedia (AH), and seeks to reduce the navigation burden upon the user by removing links that are not useful (Brusilovsky, 1996). However, AH relies upon extracting information from the user about which links are likely to be useful or not useful. However, the process of obtaining information about the user can be distracting and does not adequately reflect the user's potentially complex goals. This is the crux of adaptive hypermedia (Brusilovsky, 1996). This article investigates one possible solution to this problem, identifying the browsing patterns that a user makes as they navigate and using them to infer what the user is using the hypermedia for. Once this information has been identified it can be used in an Adaptive Hypermedia System to aid the user in their navigation. The authors discuss their prototype hypermedia system, which is used to record and attach meaning to browsing patterns with a view to employing the information in future hypermedia systems. Experiments, conducted to investigate how different types of users make different browsing patterns as they used our prototype hypermedia system, are outlined and discussed. It is argued that these experiments and supporting research give strong grounds for the use of browsing patterns as a means of obtaining information about a user without distracting them. ********** Hypermedia has long been seen as a promising medium for information storage and retrieval and as a medium for an educational system (Jonassen & Grabinger, 1991). However, a major concern with the use of hypermedia is that of user navigation. The issue is how, on the one hand, to prevent the user from becoming overwhelmed with information and losing track of where they are going, while on the other hand permitting them to make the most of the facilities that the hypermedia offers. One approach to remedy this situation is to restrict the number of links made available to the user. The concern that this might lead to an impoverished set of navigational opportunities can be countered by the use of adaptive hypermedia, which seeks to select the most useful links, from a possibly large set of links, as the user proceeds (Brusilovsky, 1996). However, a major drawback of this approach is that information must be obtained from the user, concerning their interests and level of expertise, so that the most relevant links can be presented to them. Current solutions to this problem tend to focus on questioning the user, which can be distracting (Kruschwitz, 2001). This article is concerned with a method for obtaining information about the user, as they use the hypermedia, without adversely interrupting them. Our approach attempts to obtain information about the user using two methods which we term "content-less browsing" and "content-based browsing." Content-less browsing involves an examination of the distinct navigational patterns made by the user as they navigate the hypermedia and comparing them to previous users' interactions and successes within hypermedia systems. Content-based browsing, on the other hand, attempts to determine the user's interests by examining the contents of nodes previously visited. These issues are being explored through the production of a prototype hypermedia system (Mullier, 1999). This article will give a brief overview of our model for a hypermedia system and describe and discuss experiments conducted using the prototype to explore the content-less approach to information extraction. …