Navigating the Web: Possibilities and Practicalities for Adaptive Navigational Support

John Eklund Faculty of Education PO Box 222, Lindfield NSW 2070 Australia University of Technology, Sydney Email: j.eklund@UTS.edu.au

Romain Zeiliger CNRS-IRPEACS 93, chemin des Mouilles, BP167 69131 Ecully Cedex, France Email: zeiliger@irpeacs.fr

Keywords: Adaptive Hypermedia Systems, Information Retrieval, Navigation, User-model, Web

Introduction : The educational use of the Web

The rapid growth of the hypertext-based Web makes it possible to access an immense volume of rapidly evolving information with potential application and relevance in an educational setting. In contrast to a database, this corpus of information is also a disorganised collection of both sites and documents: a so named cyberspace where "cybersurfing" has developed as a surface navigation on an ocean of information. Turning data into meaningful information is an individual process, part of any learning or information-retrieval activity, which can be supported by technology.

It has been suggested by Ibrahim and Franklin (1995) that the pedagogical use of the Web can evolve along two major axes: a closed corpus of material where the technology is used mostly for its hypermedia and distance delivery capabilities, or an open corpus approach which exploits the enormous amount of information that is accessible via Internet, whether or not it has been put there for educational purposes. An organized structure of links has then to be superimposed on the domain to allow guided educational explorations. These two axes can be alternatively or complementarily followed. We will refer to both perspectives in this paper.

Hypermedia usability research developed long before the Web arrived. New technologies - such as adaptive hypermedia systems - have been developed in response to problems observed in hypermedia use. Applying the research findings in closed corpus hypermedia to the broader domain of the Web is a logical step, and one in which navigational issues are emphasised. In this paper, we will discuss the applicability of adaptive navigational support to Web navigation.

Navigation and comprehension issues

Hypertext, due to its characteristic node-link structure, is not necessarily read in a linear manner as conventional text. By clicking on hotwords or buttons, hypertext users traverse contextual connections, a process known as navigation. This logical structure raises new issues about the model of learning on which the media is based, and the principles to be embodied in the design of hypertext-based systems in order to maximise learning.

The greatest strength and weakness of a HMS (Hyper Media System) lies in the issue of navigation (Jonassen & Mandl, 1990; Nielsen, 1990; Rivlin et al, 1994; Linard & Zeiliger, 1995, [ HREF9 ]), although the literature strongly suggests that the navigational features of HM are a positive attribute. As Jacobs (1992) notes, "The principal attraction of hypermedia is that it lends itself to naturally non-sequential educational approaches, since it encourages the free association characteristics of human thought." (Jacobs, 1992, p. 119). In the educational domain, hypermedia is perceived to offer learners complete control over the viewing of the material, within their navigational abilities. It is a cognitive tool, allowing students to explore and make sense of a knowledge corpus, "constructing" meaning in a self motivated and self directed fashion (Jonassen, 1992).

Hypermedia can allow students to learn at their own pace (Laurillard, 1993) but some direction may be necessary for hypermedia to be an effective educational tool. In fact, there is a growing body of empirical evidence to suggest that learners tend to make poor decisions in learner controlled systems ( eg Jonassen, 1990; Jonassen & Wang, 1993). Studies cited by Espinoza & Hook [ HREF3 ] have identified user's background knowledge and spatial ability as the major influences on understanding the domain and its documentation. Students become lost, skip important content, choose not to answer questions, look for visually stimulating rather than informative material, and use the navigational features unwisely. Finally, HMS are based on the assumption that the student's interpretation of the courseware is more meaningful than that of the expert or the author (Jonassen & Wang, 1990; Eklund, 1995, [ HREF8 ]). This implies that hypermedia transfers much of the mental load from the teacher to the learner.

In the information-retrieval domain, hypermedia provides opportunities for incidental learning. Users can browse along hyperlinks, explore the information space and find pieces of information which they actually never can request by a formal query (Brusilovsky, 1996). However, where the hyperspace is very large and when the system is likely to be used by people with different goals and knowledge, the risk of unproductive wandering in the link network is clearly unacceptably high. A number of writers (de La Passardiere & Dufresne, 1992; Duchastel, 1991; Jonassen, 1992; Costa Pereira et al, 1992) have suggested that an HMS with a form of expert assistance or guidance, perhaps as individualised navigational advice or help sequences would provide more structure to the information space and more direction for the user to help solve these problems of disorientation.

We now discuss what means are available to organise, or structure the hypermedial knowledge.

Knowledge structure in HMS

In this article we define knowledge structures in a HMS as being either internal or external. Internal structure refers to the representation of the content imposed on the hypermedia (HM) through the sequencing of the nodes and links, and through the provision of advanced navigational tools. It has been suggested that organising the HM using a concept map (or cognitive map) as a semi-formal means of knowledge representation (Gaines &Shaw, [ HREF4 ]; Jonassen & Wang, 1993) provides a means of internally structuring the HM according to a model of cognition for the domain ( see also Spoer, 1994). In the educational domain, the partitioning of courseware into concept-related nodes, or logical pages (Hekmatpour, 1995) in which the concepts are logically ordered and linked by an expert who imposes her understanding of how learning is sequenced in the domain, reflecting the cognitive structures of the learner, is a technique for adding internal structure to a HMS. Asking students to generate concept maps can be effective as an instructional technique in that it encourages them to organise and systematise their knowledge, identify gaps and be explicit about relationships between ideas. Further, this structural knowledge has been shown to be highly individual (Jonassen & Wang, 1993), and other studies have made the distinction between the concept maps of novices, which tend to organise the domain according to easily recognisable surface similarities such as terminology, while experts used a more hierarchical structure showing a greater number of links between similar concepts and less links between clusters of different concepts. Cognitive maps and the more formalised semantic networks (Quillian, 1968) in AI are a form of visual language for the representation of knowledge in, and internal structure of, a HMS.

Whether the knowledge structure is imposed on the hypermedia in an informal way or developed through a cognitively based theory (Recker et al, 1995), the environment remains passive, ignorant of the individual knowledge state of the user. This is overwhelmingly the case with the Web. Further, the results of the empirical studies described above underscore weaknesses in learning from internally or semantically structured HM, and strengthen the case for a more dynamic external structure to be applied to the HM which accounts for the specific knowledge and tasks of an individual user.

External structure in a HMS refers to the expert knowledge being embedded to the extent that the HMS is capable of dynamically altering content or links to suit the needs of the individual user (or more precisely, to suit the expert's view of the needs of the individual user). This is achieved through the use of some form of knowledge representation, both of content and in the user-model (UM), which is the system's understanding of the user's current knowledge state. This is a fundamental principle of the generalised ITS architecture.

Adaptive Hypermedia Systems

Adaptive hypermedia systems (AHS) instantiate a relatively recent area of research integrating two distinct technologies in computer assisted instruction, Intelligent Tutoring Systems (ITS) and Hyper Media Systems (HMS). ITS are knowledge centred and have the ability to individualise instructional sequences by modifying content or presentation based on the interactions with the student. HMS are predominantly a non-pedagogical technology (Duchastel, 1992) in which learning relies on the user's interest and purpose through the use of a variety of navigational aids in a database of hyper linked information. Integrating the two technologies (Jonassen & Wang, 1990; Costa Pereria et al, 1991; Duchastel, 1991; Douglas, 1994) is in effect a combination of two opposed approaches to computer assisted learning : the more directive tutor-centred style of the traditional AI based systems, and the flexible student-centred browsing approach of a HMS.

AHS are defined as externally structured systems which are explicitly based on hypertext (or hypermedia), and use a model of the user's knowledge or goals to modify links or content to present individualised instruction or guidance. They have been applied to such diverse domains as writing (Carlson & Gonzalez, 1993), hospital administration (Vassileva, 1994), the C programming language (Kay & Kummerfield, 1994 [ HREF2 ]), anatomy (ANATOM-TUTOR, Beaumont, 1994, [ HREF6] ), and statistics (deRosis et al, 1993). Research in this area ( see The Adaptive Hypertext and Hypermedia Home Page, [ HREF5 ]) is relatively recent and interest in the educational and technical aspects of AHS is growing rapidly.

Brusilovsky [ HREF1 ] provides two main categories of features which can be dynamically adapted in an AHS: Adaptive presentation and adaptive navigational support.

Adaptive presentation works at content-level, the information contained in the hypermedia nodes (or pages) can be presented in an adaptive fashion, to vary the detail, explanation, or media use (text, graphics, sound), or to incorporate more or less link-anchors (hotwords in hypertext) . Most current work in this category focuses on text adaptation. This style of adaptation addresses the problem of a HMS being used by different classes of users by tuning the information presentation. One method of achieving this is through limiting the browsing space to meet users with different needs, background knowledge, interaction style and cognitive characteristics (Kosba, 1994; Recker et al, 1995). Systems that use adaptive presentation make underlying assumptions about users. For example, the adaptive presentation model for hypermedia of Hekmatpour (1995) uses adaptive re-ordering of node-objects using the frequency of visits to each object as a measure of its usefulness to the user. This model assumes that a relevant node will be frequently visited, when there may be many other reasons why some nodes are visited often.

The second category of adaption is adaptive navigational support which works at link-level (or more precisely we could say at "path-level"). In this mode, the space of possible paths which can be followed by users may be tailored. This style of adaptation would address the problem of users being lost in hyperspace by providing guidance and orientation support. It should be noted however that those two styles of adaptation are closely related because most links have their anchors in content presentation and because the fragmentation of information presentation is done by way of node-link development. For example a more detailed explanation can be provided to the user either by expanding the text on the current page or by adding more in-depth links. Adaptive interfaces (eg Pitz, 1994) are a further combination of the two categories, where the appearance and form of the page and the navigational devices present change with the user's knowledge or tasks, and this indirectly affects the content that can be viewed in the system. This cutomisation of user interfaces is an accepted technique (Benyon et al, 1994) to improve a system's usability.

In this paper we wish to focus on the aspect of adaptive navigational support in a HMS and discuss actual and possible applications to the Web. We will now consider briefly the question of what are the user's features on which adaptation can be based.

The Role of the User Model

Individualising information and link-anchors (adaptive presentation) or providing the user with navigational support (adaptive navigation) is performed within a system on the basis of information kept in a user-model, the system's representation of the user's preferences, knowledge, beliefs, or information seeking goals. Paiva, Self & Hartley (1995) have succinctly defined the nature and purpose of a student model (or UM for User Model) as "...representations of some characteristics and attitudes of the learners, which are useful for achieving the adequate and individualised interaction established between computational environments and students. They are constituted by descriptions of what is considered relevant about the actual knowledge of a learner, providing information for the learning environment to adapt itself to the individual learner." (Paiva, Self & Hartley, 1995, p.509). For instance, in the ISIS-Tutor, an AHS dedicated to the teaching of some aspects of programming language and providing adaptive navigational support (Brusilovsky & Pesin, 1994; Beaumont & Brusilovsky, 1995) the student model is an account of whether, for a particular student, a node is learned, ready to be learned, or unlearned. According to Brusilovsky, in adaptive hypermedia systems five main user's features are represented inside user models, either alone or in combination. These are:

Different techniques have been used to acquire information about the user (Kobsa, 1994). These include:

The effectiveness of some of these techniques remains controversial. Judy Kay notes that "Modeling cannot be anything but a guess if it attempts to model the user's knowledge" (Kay, 1994). She and others (Hook et al, 1995) advocate that a user model should be "glass box", meaning inspectable by users and should work under their control.

In the closed corpus educational system scenario the tutor has a significantly greater opportunity to obtain information about the user through testing. The HM may be arranged in a hierarchy of interrelated concept nodes, each node consisting of multiple pages of material including explanations, remediation, extension, review and testing. It is naturally possible to mount such systems on the Web as a cluster of self-contained nodes, and incorporate a variety of plug-ins for interactivity and data collection as found in the PUSH system (Espinoza and Hook, [HREF3]), described later in this article. Information gained from testing enhances the quality of the data held in the UM and this is commensurate with its ability to customise information or links.

While the development of some AHS is based on a recognition of user's comprehension problems (Kobsa, Muller & Nill, 1994), adaptive navigation is particularly useful in larger HM environments, and has been the motivating force in its implementation in a number of others (Boecker et al, 1990; Kaplan et al, 1994). Given the open-corpus scenario of the Web, where the emphasis is information seeking (Hook, 1995, [HREF10]) and navigation rather than learning in a narrowly restricted domain, a user model (UM) might contain an account of the navigational characteristics of a user to provide adaptive navigational support.

For information retrieval systems, the UM may function through stereotyping the user (Vassileva, 1994; Hook et al, 1995) or through the user's own indication of the relevance of the information (Mathe & Chen, 1994). The stereotypical knowledge is derived from observations of groups of users, who can "...seriously differ in their goals, background and knowledge on the subject covered in the HMS." (Brusilovsky, 1994, [HREF1]). This is a technique that avoids the need for the very complex and detailed knowledge representation as is the case in an ITS. Whatever architecture is employed, there is clearly a need to base the stereotyping of users on an empirical grounding. In a study of Web navigation strategies, Catledge and Pitkow [HREF7] propose a characterisation of users which goes beyond the browser/searcher distinction and may provide a basis for more fine-grained navigational adaptation. These are:
- Serendipitous Browser These users avoid the repetition of long invocation sequences. This shallow browser may be reflective of a Web repository structuring in that the databases visited by these users may be weakly connected.
- General Purpose Browser Here users perform as expected. Probabilistically, they have roughly a one in four chance of repeating a more complex navigation sequence. This is the average inertia for all users sampled.
- Searcher A user performs the same short navigation sequences relatively infrequently, but does perform long navigational sequences often.

Other work in establishing the user-centred principles upon which a navigational support system may be designed (Linard and Zeiliger, 1995) recognises that knowledge of content and interface familiarity are key components in determining a level of support. This work informs the design of the user-model in an AHS. Linard & Zeiliger (1995) suggest a three-phase model for educational software which takes the learner through an introduction, a tutorial and finally free-browsing with adaptive navigational support. A teaching architecture in which the user increases their level of automomy with time, progressing to more active learning through orientation, coaching and tuning stages has similarly been suggested in the ITS literature, and is based on cognitive theory (Chan, 1993).

Brusilovsky (1996, in press) writes about personalized views in world-wide information spaces as a means of managing vast information resources and tailoring the hyperspace to the information seeking tasks of the user. The idea of individual users defining these views to "...protect themselves from the complexity of the overall hyperspace" (Brusilovsky, 1996, in press) is based on the notion that users have specific information retrieval tasks, which are concrete enough to define and maintain. In the task-based domain of information retrieval in an institution such as a hospital, this seems feasible. In the broader educational domain, however, the problem is that a single user may have different goals, tasks and methods at various times, and so may require multiple user models. In non-institutional information retrieval, such as browsing the Web for information related to a specific topic, we need to distinguish a user's specific and general interests. This is because a user's information seeking goals can vary markedly. For example, I may spend some time using a number of search tools to find material related to my topic of interest, currently adaptive hypermedia. My adaptive browser records the keywords and notes the nodes I visit and those that I have indicated as relevant, records my time on them, and begins to sort the hits from successive searches in order of the likelihood of them being useful to me. Later I may instruct the browser to continue searching for related, relevant nodes. Before I do, however, my interest wanders and I open a bookmark to a weather map of Sweden. If I do not tell my browser that this node is not related to the main information seeking goal I will confuse it. Two options present themselves, both related to the notion of a collaborative user-model, that is, one into which both the user and the computer have input: to allow the user to distinguish between tasks to inform the user model if some item is relevant to a task, or to simultaneously run multiple models of a user's task.

Clearly there are multiple influences on the design and implementation of the adaptive systems: Studies of pedagogy and their interpretation as architectures for computer-based learning environments, learning and cognitive theory, instructional design principles, developments in software tools and methods, and empirical studies about the nature of the user and the effectiveness of the environment. A successful integration of a number of these areas is required to form guiding principles to inform the research area of AHS.

Methods and techniques for navigation support

Navigational adaptivity may be implemented in a number of ways (Brusilovsky, 1995): direct guidance, adaptive hiding or re-ordering of links, adaptive annotation, or map adaptation.

A simple adaptation mechanism is provided by most Web browsers in highlighting the links to previously visited nodes: even this simplest form of annotation appears to be quite useful. The general idea of individualised annotation is to augment the links with some form of comments which can tell the user something about the document behind the link that could be in relation with his current task. Annotations can be provided in textual form (Zhao et al, 1993) or as visual cues (Brusilovsky, 1994).

Direct guidance is a technique with its roots in ITS research: on the basis of the user model, the system decides what is the best next node for user's visit; the "best next link" can be either one among those of the current page (and then it can be outlined) or a dynamically generated link, usually a "next" button, which is added as a complement of the current page. In contrast with some ITS approaches, the user is usually free to decide whether to follow the system's suggestion. Guidance is provided here as a dynamically-computed additional path which therefore broadens the hyperspace.

The link re-ordering technique sorts the available links on the basis of information contained in the user model, displaying the most relevant links on top. This technique applies only to non-contextual links and not hotwords. Hiding the less-relevant links is the most often used technique, and applies to all links, contextual or not. It supports navigation by restricting the navigation space. It has been criticised for it may induce building of incorrect mental models of the knowledge domain.

In the following section we examine systems which are more closely related to the Web.

Applying the lessons from AHS to the Web

HyperText Markup Language (HTML) has been successfully integrated into the interface of a number of AHS. Kay and Kummerfield (1994, [HREF2]) describe an implementation of an architecture for delivering an individualised course in the C programming language, based on HTML. The authors define a metahypertext which allows pages, links and the semantic structure to be dynamic. The C course consists of component HTML pages which are customised for the individual user. This work provides' customised hyperspaces. The user model in this system is glass-box and user-authorable, and based on modifications to stereotypical 'views', each of which are a set of structural relationships about a user's knowledge. In the case of users learning C, the other programming languages in which they have background has been shown to make major differences.

Espinoza and Hook [HREF3] deal with the problem of information overload to make the Web more interactive through the modular PUSH (Plan and User Sensitive Help) system which has the capability of generating Web pages dynamically. The system utilises the user's information seeking task as a basis for an adaptive filtering mechanism. The PUSH system is a platform independent, interactive, adaptive system for filtered information retrieval on the Web. The up-datable knowledge database and the interface are separated in the architecture [HREF11], in which the student model is based on the information seeking tasks of the user.

Most adaptive systems in the closed-corpus approach rely on a dynamic generation of HTML pages (as distinct from stored pages). This mechanism is supported by Web servers via Common Gateway Interface. In The Moscow University project (Brusilovsky, 1995) courseware generated by the system is a combination of HTML files, knowledge frames and cgi scripts. HTML pages are adaptively generated from the knowledge base by the scripts attached; links are individually "provided with comments". The goal of this project is to design an authoring shell which can be used by a human tutor to design adaptive courseware for distance education. The first field of application is an "introduction to UNIX" which can be used as a UNIX adaptive reference manual.

Brusilovsky (1996, in press) defines global guidance in a hypermedia system as an important adaption goal for users who require less-specific information from a number of nodes that may be obtained through browsing, with navigation support being a technique to attain this goal. WebWatcher (Armstrong et al., 1995) is an excellent example of a navigation support technique implemented for users of information an retrieval system, namely the Web. This system uses a direct method of suggesting at each step which of the links to follow.The Web Watcher Home Page [HREF12] describes WebWatcher as a "tour guide agent for the Web. Once you tell it what kind of information you seek, it accompanies you as you browse the web, highlighting hyperlinks that it believes will be of interest. Its strategy for giving advice is learned from feedback from earlier tours."

To attempt to generalise about the applicability of adaptive navigational systems to the Web in terms of the closed and open corpus perspectives on which our discussion is based, let us consider possibilities for each scenario. For closed systems, the Web component (residing on the server) is likely to be either a generated HTML interface between the student and courseware which provides a range of appropriate navigational tools, or a front-end to an adaptive information retrieval system. For navigation and information retrieval on the Web, the so-called open-corpus scenario, we think that having an adaptive module (residing on the client in the form of a local JAVA applet for example) to record the users interests, frequency of visits to links, characteristic uses of navigational tools and information seeking tasks, and to adaptively annotate links, is the most probable application to the Web of the lessons from adaptive hypermedia.

Conclusion and directions for further research

In this paper we have outlined the relatively new research field of AHS as promising to provide increased opportunities for individualised learning in courseware and information retrieval systems. AHS are characteristically "personal helpers" who guide rather than direct, and can be used to alleviate the navigational problems that users of HMS experience. HTML has been successfully integrated into the interface of a number of these AHS and we have discussed their application to the Web. In the open-corpus scenario of the Web, particularly for information retrieval, user modeling may be based on characterizations of user's navigational traits. Further empirical studies on the navigational use of the Web will form the basis for forming classes of users to implement adaptive navigational support systems, helping users orient themselves in the hyperspace. There is a clear need within the literature of AHS for empirical studies, both in the analysis of user's navigational traits in HM to inform the principles upon which adaptation takes place (i.e. the design of the user-model), and in the subsequent evaluation of the systems themselves as viable educational tools.

The relatively recent tremendous interest and acceptance of the Web offers us the ideal vehicle to implement distance education and modes of alternate delivery of educational material. Within the community of educators the Web has been seen as an information resource and a tool for collaboration and publication and on this basis it has been enthusiastically integrated into classrooms (eg Journal of Computers in Mathematics and Science, 1995 Vol 14 No 1/2; Nott et al, 1995), but its full value as a teaching medium for individual instruction has yet to be realised. Kay and Kummerfield (1994, [HREF2]) note that "Hypertext Systems such as the World Wide Web hold great promise as a vehicle for delivering self-paced instructional material". Its first great step to prominence came with the application of a hypertext based graphical user interface to the Internet. Giving the Web intelligence: the ability to "understand" the user, to customise information and presentation and to dynamically support navigation; may be the next significant leap from a popular medium to one which is a valuable teaching mechanism in its own right. In the near future we may be able to download a new Web-browser plug-in which offers adaptive navigational support based on stereotypical user's navigational traits, as a mechanism to support learning and information retrieval from the Web.

References

Armstrong R, Freitag D, Joachims T, & Mitchell T (1995) WebWatcher: A learning apprentice for the World Wide Web. Proceedings of AAAI Spring Symposium on Information Gathering from Distributed, Heterogeneous Environments

Beaumont I & Brusilovsky P (1995) Adaptive hypermedia: From ideas to real systems. In Hermann Maurer (ed.) Educational Multimedia and Hypermedia, 1995. Proceedings of ED-MEDIA95, p. 93-98.

Boecker H, Hohl H & Schwab T (1990) Hypadapter-Individualizing Hypertext. In Diaper D et al (ed.) INTERACT '90. Proceedings of the the IFIP TC13 Third InternationalConference on Human-Computer Interaction . North-Holland, Amsterdam. 931-936.

Benyon D, Kuhmle T, Malinowski U & Sukaviriya P (1994) Computer-aided adaption of user interfaces. SIGCHI Bulletin. Vol 26 No 1 p. 25-27.

Brusilovsky P (1996, in press) Methods and techniques of adaptive hypermedia. Paper to be presented at UMUAI.

Brusilovsky P & Pesin L (1994) ISIS-Tutor: An adaptive hypertextlearning environment. Proc. of Japanese-CIS Workshop on knowledge-based software engineering. Tokyo.

Carlson P & Gonzales G (1993) The knowledge board: Using hypertext as an "intelligent" workspace for writing issues-based prose. Journal of educational multimedia and hypermedia. Vol 2 No.14, p 417-431.

Chan T, Lin C, Lin S and Kuo H (1993) OCTR: A model of learning stages. In Brna P, Ohlsson S and Pain H (eds) Proceedings of AI-ED 93 . Scotland, August 1993. ACCE publication.

Costa Pereira D, Oliveira A & Vaz J (1991) Hypermedia and ITS. Oliveria A (ed.) Hypermedia courseware: Structures of communication and intelligent help. Springer-Verlag.

de La Passardiere B & Dufresne A (1992) Adaptive navigational tools for educational hypermedia. In Tomek I. (ed.) Computer Assisted Learning. Proceedings of the 4th International Conference,ICCAL'92. Springer-Verlag, Berlin. 555-567.

Douglas I (1994) Combining and expert system and a hypermedia system into a tutoriang system. Proceedings of AI-ED93. p 551.

de Rosis F, De Carolis B, & Pizzutilo S (1993). User tailored hypermedia explanations. Proceedings of INTERCHI'93 (pp. 169-170). Amsterdam.

Duchastel P (1991) Integrating hypermedia into intelligent tutoring. Oliveria A (ed.) Hypermedia courseware: Structures ofcommunication and intelligent help. Springer Verlag, Berlin.

Duchastel P (1992) Towards Methodologies forBuilding Knowledge-Based Instructional Systems. Instructional Science,v20 n5-6 p349-58.

Eklund J (1995) Cognitive models for structuring hypermedia and implications for learning from the world-wide web. In Debreceny RS and AE Ellis, Ed. (1995). Innovation and Diversity - The World Wide Web in Australia. AusWeb95 - Proceedings of the First Australian World Wide Web Conference. Lismore, NSW, Norsearch Publishing. p.111-116.

Hekmatpour A (1995) An adaptive prsenetation model for hypermedia information systems. Journal of Educational Multimedia and Hypermedia. Vol 4 No 2/3 p 211-238.

Hook K, Karlgren J & Waern A (1995) A glass box intelligent help interface. First workshop on intelligent multimodal interfaces, Edinburgh, 1995.

Hook K (1995) Adapting explanations to the user's task. SICS Research Report R95008, SICS, Sweden, 1995.

Ibrahim B & Franklin S (1995) Advanced uses of the World-Wide Web. Computer Networks and ISDN systems . No 27 (1995), p. 871-877.

Jacobs G (1992) Hypermedia and dioscovery-based learning: ahistorical perspective. British Journal of Educational technology. Vol23 No 2 p 113-121.

Jonassen D & Grabinger R (1990) Problems and issues indesigning hypertext/hypermedia for Learning in Jonassen D and Mandl H (eds) Designing Hypermedia for Learning. Singer-Verlag. Berlin.

Jonassen D & Mandl H (eds) (1990) Designing hypermedia for learning. Springer-Verlag, Berlin.

Jonassen D (1992) Designing Hypertext for Learning. In Scanlon E & O'Shea T (eds) New directions in educational technology. Springer-Verlag. Berlin.

Jonassen D & Wang S (1990) Adding intelligence to hypertext with expert systems and adding usability to expert systems with hypertext. In Oliveria A (ed.) Hypermedia courseware: Structures of communication and intelligent help . Springer Verlag, Berlin.

Jonassen D & Wang S (1993) Acquiring structural knowledge from semantically structured hypertext. Journal of Computer-based Instruction. Vol 20 No 1, p 1-8.

Kaplan C, Fenwick J & Chen J (1993) Adaptive hypertext navigation based on user goals and context. User models and user adapted interaction, 3(2).

Kobsa A, M ller D & Nill A (1994) KN-AHS:AN adaptive hypertext client of the user modeling system BGP-MS. Fourth International Conference on User Modeling, Hyannis, Massachusetts, USA, August 1994. Mitre Corporation.

Laurillard D (1993) Rethinking University Teaching, a Framework for the
effective use of Educational Technology, Routledge, London.

Nielsen J (1990) Hypertext and Hypermedia. Academic press, San Diego.

Nott M, Riddle M & Pearce J (1995) Enhancing traditional university science teaching using the world wide web. In Selwood I, Fox P and Tebbutt M (eds.) Conference Abstracts of WCCE95, the Sixth IFIP World Conference Computers in Education.

Paiva A, Self J & Hartley R (1995) Externalising LearnerModels. In J Greer (ed.) Proceedings of AIED95. AACE publication.

Pitz H (1994) "Adaptive customisation" of user interfaces. Fourth International Conference on User Modeling, Hyannis, Massachusetts, USA, August 1994. Mitre Corporation.

Rivlin E, Botafogo R & Schneiderman B (1994) Navigating in hyperspace: Designing a structure-based toolbox. Communications of the ACM. Vol 37 No 2, p87-96.

Vassileva J (1994) A Practical architecture foruser-modeling in a hypermeia-based information system. FourthInternational Conference on User Modeling, Hyannis, Massachusetts, USA,August 1994. Mitre Corporation.

Quillian, M (1968). Semantic memory. Minsky, M (Ed.) Semantic Information
Processing. p. 216-270. Cambridge, Massachusetts, MIT Press.

Recker M, Ram A, Shikano T, Li G & Stasko J (1995) Cognitive media types for multimedia information access. Journal of Educational Multimedia and Hypermedia. Vol 4 No 2/3 p 183-210.

Spoer (1994) Enhancing the acquisition of conceptual structures through hypermedia. In K McGilly (ed.) Classroom lessons: Integrating Cognitive theory and classroom practice. Bradfort MIT Press, London.

Zao Z, O'Shea T & Fung P (1993) Visualisation of semantic relations in hypertext systems. In T. Ottman & I. Tomek (eds.) Educational multimedia and hypermedia. Proceedings of Ed-Media 93, AACE pp 556-564.

Hypertext References

HREF1: http://www.edfac.usyd.edu.au/projects/ah/Brusilovsky.html - Adaptive Hypermedia: an attempt to analyse and generalise.
HREF2: http://www.ncsa.uiuc.edu/SDG/IT94/Proceedings/Educ/kummerfeld/kummerfeld.html - Kay J & Kummerfeld R (1994) An individualised course for the C programming language.
HREF3: http://www.sics.se/~kia/espinoza_hook.html - Espinoza F and Hook K (1996, in press) An interactiveWWW interface to an adaptive information system. Paper to be presented at UM'96
HREF4: http://ksi.cpsc.ucalgary.ca/articles/ConceptMaps/ - Gaines B & Shaw M (1995) Concept Maps as Hypermedia Components.
HREF5: http://www.edfac.usyd.edu.au/projects/ah - The Adaptive Hypertext and Hypermedia Home Page.
HREF6: http://www.edfac.usyd.edu.au/projects/ah/Beaumont.html - Beaumont I (1994) User modelling and hypertext adaption in the tutoring system Anatom-tutor.
HREF7: http://www.gatech.edu/lcc/idt/Students/Catledge/browsing/user_1.html - Catledge, Pitkow Characterising Browsing Strategies on the Web.
HREF8: http://www.scu.edu.au/ausweb95/papers/hypertext/eklund/index.html - Eklund J (1995) Cognitive models for structuring hypermedia and implications for learning from the world-wide web.
HREF9: http://www.irpeacs.fr/papers/rz/artmosc.htm - Linard M & Zeiliger R (1995) Designing navigational support for educational software.
HREF10: http://www.sics.se/~kia/espinoza_hook.htm - Hook K (1995) Adapting explanations to the user's task. SICS Research Report R95008, SICS, Sweden.(postscript file)
HREF11: http://sics.se/~kia/architecture.GIF - The Architecture of the PUSH system.
HREF12: http://www.cs.cmu.edu/afs/cs.cmu.edu/project/theo-6/web-agent/www/project-home.html - Web Watcher Home Page.

Copyright

John Eklund, Romain Zeiliger ©, 1996. The authors assigns to Southern Cross University and other educational and non-profit institutions a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The authors also grant a non-exclusive licence to Southern Cross University to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the conference papers, and for the document to be published on mirrors on the World Wide Web. Any other usage is prohibited without the express permission of the author.

Pointers to Abstract

Abstract

Papers & posters in this theme

All Papers & posters

AusWeb96 Home Page

AusWeb96 The Second Australian WorldWideWeb Conferenceausweb96@scu.edu.au