1 KN-AHS: An Adaptive Hypertext Client of the User Modeling System BGP-MS Alfred Kobsa, Dietmar Müller, Andreas Nill Dept. of Information Science, Univ. of Konstanz P.O. Box 5560-D73, 78434 Konstanz, Germany {kobsa,mueller,nill}@inf-wiss.uni-konstanz.de Abstract 1 This paper describes the automatic adaptation of hypertext to the user’s presumed domain knowledge in the KN-AHS system, and the support that the user modeling shell system BGP-MS can provide for this adaptation. First, basic hypertext concepts will be introduced and reasons given for why hypertext should adapt to the current user (especially to his/her state of knowledge). A brief overview of those representation and inference components of BGP-MS that are used by KN-AHS will then be provided, followed by a description of its adaptive user interface. The interaction between the adaptive hypertext system and the user modeling system will be investigated in detail based on a possible dialog with a user. Finally, the inter-process communication between KN-AHS and BGP-MS will be described and related work discussed. The aim of this work was to demonstrate the feasibility of user modeling with BGP-MS in a “normal” hardware and software environment that is frequently found in the workplace. 1. Hypertext and adaptive hypertext 1.1 Hypertext Hypertext consists of any number of objects 2 that can be linked with one another in a network structure. Therefore, hypertext is not necessarily read in linear (i.e., sequential) order like conventional text, but can be read in a non-linear order by navigating within the hypertext node network. This non-linear linkage of objects represents the basic characteristic of hypertext [cf. Seyer 1991, Kuhlen 1991]. Gaining information in a non-linear form is not new (see e.g. information search in encyclopedias). As opposed to printed media, however, the representation of information in electronic form allows the user to directly and comfortably traverse contextual connections. Hypertext has therefore been able to enjoy an increased importance in the last few years, especially as a basis for on-line help systems and electronic encyclopedias. The user-friendliness of many hypertext systems is rooted in their usage of intuitively understandable direct- manipulative interfaces [cf. Kuhlen 1991, Shneiderman & Kearsly 1989]. The user has the possibility of directly manipulating graphical objects with a pointer (e.g. a mouse) 1 This work was supported by the German Science Foundation (Grant Ko-1044) and the Univ. of Konstanz (Grant AFF 17/92). 2 The objects in a hypertext base are not only text documents but can also include non-textual data (tables, graphs, animation, etc.). If an audiovisual component is involved, then the term ‘hyper- media’ is used. without having to use complicated commands. The effect of these actions can be seen immediately on the screen. Direct manipulation can be used, for example, to reach a different node from the node currently shown on the screen. The usual graphical objects used in such navigating operations include mouse-sensitive text passages (hotwords) or buttons (more on this in Section 3). Other important hypertext components are glossaries, indices, and graphical representations of the hypertext structure, which all offer important meta-information about the basic text objects. Recent research supplements associative navigation by controlled navigation and by search techniques from the field of Information Retrieval, in order to increase the search efficiency [cf. Kuhlen 1991]. 1.2 Adaptive Hypertext Two major problems arise when working with hypertext systems:  Orientation and navigation problems: When navigating in hypertext, users are frequently uncertain as to how to reach their goals. Since users can choose any course within hypertext, they run the risk of losing their orientation. Navigation aids that take users’ goals into account may be helpful. Kaplan et al. [1993] showed empirically that navigation suggestions based on knowledge about the objectives and the navigation behavior of previous users as well as the goals of the current user can significantly accelerate the current user’s search for information.  Comprehension problems: Since hypertext is frequently read by users with differing knowledge and experience levels, it may at the same time be too difficult and too detailed for laypersons, and too redundant for experts. Boyle & Encarnacion [1993] showed empirically that an automatic adaptation of hypertext to the user's state of knowledge significantly improves text understanding as well as partially improving search speed. The system KN-AHS 3 deals with the second problem and adapts hypertext objects to the current user's state of knowledge. In contrast to other adaptive hypertext systems, the realization of KN-AHS took advantage of existing software products. TOOLBOOK [Asymetrix 1989], a widely available hypertext shell system, offered us a powerful tool for the implementation of the hypertext and its user interface. The user modeling shell system BGP-MS offered a wide variety of representation and inference possibilities that ensure flexible adaptation. Both tools run as independent software systems on a PC platform and interact via inter-process communication. 3 KN-AHS stands for KoNstanz Adaptive Hypertext System. Proceedings of the Fourth International Conference on User Modeling, Hyannis, MA, 99-105, 1994.