Adaptive hypertext navigation based on user goals and context

Abstract

Abstract. Hypertext systems allow flexible access to topics of information, but this flexibility has disadvantages. Users often become lost or overwhelmed by choices. An adaptive hypertext system can overcome these disadvantages by recommending information to users based on their specific information needs and preferences. Simple associative matrices provide an effective way of capturing these user preferences. Because the matrices are easily updated, they support the kind of dynamic learning required in an adaptive system. HYPERFLEX, a prototype of an adaptive hypertext system that learns, is described. Informal studies with HYPERFLEX clarify the circumstances under which adaptive systems are likely to be useful, and suggest that HYPERFLEX can reduce time spent searching for information by up to 40%. Moreover, these benefits can be obtained with relatively little effort on the part of hypertext authors or users.The simple models underlying HYPERFLEX's performance may offer a general and useful alternative to more sophisticated modelling techniques. Conditions under which these models, and similar adaptation techniques, might be most useful are discussed.

A task-centered approach for user modeling in a hypermedia office documentation system

“…However, we would have needed users who are confident working with the system and who give reliable feedback. Novice users are not good teachers for a machine learning system.Other ways of getting feedback, for example, using the assumption that the more time is spent on an unit, the more interesting it is (Kaplan et al, 1993), can also be unreliable, since it is not clear what the user is actually doing in this time. Of course, if she spends too little time, not enough even to read the information, this is a sort of evidence that the user was not interested in it.…”

“…It is based on weighted links between the tasks and the hypermedia entities. Unlike the normalized weights of Hyperflex (Kaplan et al, 1993) which denote the strength of association between a goal and a node, we use discrete values:-2 -forbidden for the user entity, display blocked from any task; -1 -"forbidden" for the task entity, display blocked for current task; 0 -neutral entity, not related to the task, but accessible through free browsing;1 -connected to the task with "read only" rights of access;2 -connected to the task with "read and write" rights of access.The two types of viewing are implemented in the following way:• free browsing with an anchor -all the entities irrelevant for the task are considered by default as connected with links of value 0. They are not included in the task view, but the user can reach them by free browsing following the standard hypermedia links.…”

“…More relevant work on this problem exists in the field of adaptive information retrieval. The semantic indexing of information with respect to the individual user's interests, goals or tasks has been extensively investigated (Belew 1986; 1 A similar structure for indexing hypermedia entities with respect to the user's goals and preferences appears later in (Kaplan et al, 1993) and in (Mathé & Chen, 1996).30 Belkin et al 1987;Kok 1991). Though using different approaches and representation schemes: blackboard architecture with cooperating experts (Belkin et al, 1987), rule based expert systems (Brajnik et al, 1990), first order logic (Kok, 1991), connectionist schemes (Belew 1986;Biennier et al 1990;Crouch et al 1994), relevance networks (Mathé and Chen, 1996), or agent-based architectures (Thomas & Fischer, 1996), these approaches share one feature: users' feedback is used as an information source for updating the semantic indexing scheme.…”

Adaptive Hypermedia for Education and Training

“…When the user searches for information, the system adaptively selects and prioritizes the most relevant items (Brajnik, Guida & Tasso, 1987;Brusilovsky, 1992b). Adaptive navigation support was introduced in early adaptive hypermedia systems (de La Passardiere & Dufresne, 1992;Kaplan, Fenwick & Chen, 1993) and is specifically associated with browsing-based access to information. When the user navigates from one item to another, the system can manipulate the links (e.g., hide, sort, annotate) to guide the user adaptively to the most relevant information items.…”

E-healthcare via Customized Information Services: Addressing the Need for Factually Consistent Information

“…Collateral structure adaptation [15], link sorting [17], link annotation [18], and link removal/addition [19] are some of the methods used. Our work does not involve any structure adaptation, as this kind of adaptation is relevant to an existing hypermedia document/website.…”