Adaptive educational hypermedia

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Adaptive educational hypermedia is one of the first and most popular kinds of adaptive hypermedia. It applies adaptive hypermedia to the domain of education. Example systems of adaptive educational hypermedia are ELM-ART[1] by Gerhard Weber et al., InterBook[2] by Peter Brusilovsky et al., Personal Reader by Nicola Henze et al., and AHA![3] by Paul de Bra et al. In contrast to traditional e-learning/electronic learning (and face-to-face education) systems, whereby all learners are offered or even directed a standard series of hyperlinks, adaptive educational hypermedia tailors what the learner sees to that learner's goals, abilities, needs, interests, and knowledge of the subject, by providing hyperlinks that are most relevant to the user. Essentially, the teaching tools "adapt" to the learner. Of course, this requires the system to be able to effectively infer the learner's needs and desires.

Many fields of research including human-computer interaction, educational technology, cognitive science, intelligent tutoring systems and computer engineering are contributing to the development of adaptive hypermedia. Unlike intelligent tutoring systems, however, adaptive educational hypermedia does not target stand-alone systems, but hypermedia systems. Moreover, the use of adaptive hypermedia is not limited to formal (or informal) education or training endeavours. Such systems can, e.g., increase profits by adapting to consumers' searches (sometimes unconscious) for goods, services, and experiences. Thus, systems like Amazon are also examples of adaptive hypermedia, recommending books based on user preferences and prior history. Other application fields of adaptive hypermedia, beside of adaptive e-learning and adaptive e-commerce applications can be adaptive e-government applications. Generally speaking, adaptive hypermedia systems can be useful anywhere where hypertext and hypermedia is used. The most popular adaptive hypermedia systems are web-based systems.

An interesting aspect of adaptive hypermedia is that it makes distinction between adaptation (system-driven personalization and modifications) and adaptability (user-driven personalization and modifications). One way of looking at it is that adaptation is automatic, whereas adaptability is not. From an epistemic point of view, adaptation can be described as analytic, a-priori, whereas adaptability is synthetic, a-posteriori. In other words, any adaptable system, as it 'contains' a human, is by default 'intelligent', whereas an adaptive system that presents 'intelligence' is more surprising and thus more interesting. This conforms with the general preference of the adaptive hypermedia research community, which considers adaptation more interesting. However, the truth of adaptive hypermedia systems is somewhere in the middle, combining and balancing adaptation and adaptability.


A common method or framework of design for an adaptive educational system is with four inter-dependent components.

  • Domain Model - The domain model contains several concepts that stand as the backbone for the content of the system. Each concept has a set of topics. Topics represent individual pieces of knowledge for each domain and the size of each topic varies in relation to the particular domain. Additionally, topics are linked to each other forming a semantic network. This network is the structure of the knowledge domain.[4][5]
  • Student Model - The student model consists of a personal profile (which includes static data, e.g., name and password), cognitive profile (adaptable data such as preferences), and a student knowledge profile.[4][5]
  • Content model - The content model describes educational content (information pages, examples, problems, etc.) in terms of Domain Model concepts. The simplest content model relates every content item to exactly one domain concept (in this model, this concept is frequently referred to as domain topic. More advanced content models used multi-concept indexing for each content item and sometimes uses roles to express the nature of item-concept relationship.[5]
  • Adaptation Module - The adaptation module displays information to the user based on his or her cognitive preferences. For instance, the module will divide a page's content into chunks with conditions set to only display to certain users or preparing two variants of a single concept page with a similar condition.[4]

The system architecture utilizes the three models in tandem. The conditional elements of these modules use simple variables representing how much to weight certain content(i.e. The likelihood of a piece of content showing up) depending on the user's preferences and history. Through use of the user model, a user can see basic concepts without any restriction. After seeing the basic concepts, a variable updates and allows the user to view advanced concepts in relation to the concept they have just seen. Additionally, as the user peruses more content, the system keeps track of the variables to build and adjust a knowledge level variable. As the user is deemed knowledgeable enough by the conditions, links and concepts considered too easy will be removed to streamline and simplify the experience for the user.

Case Studies[edit]

Aristotle University[edit]

Conducted by Evangelos Triantafillou, Andreas Pomportsis, and Stavros Demetriadis of Aristotle University of Thessaloniki, Greece. As Triantafillou (2003) states, the objective of this study is "an attempt to examine some of the critical variables, which may be important in the design of an adaptive hypermedia system based on student’s cognitive style." (p. 4) Their system offers users two modes of control to users and users have the option of changing between the two at any time. Learner control mode allows the user to view the entire hierarchy of concepts and topics and move freely through the course via links in the left sidebar whereas program control mode is automated by the system by only allowing the user to move forwards or backwards in the course's content structure.[4]

In the first two phases of the study, an expert review and one-to-one evaluation was conducted. The expert review consisted of "a teaching/training expert, an instructional design expert, a subject-matter expert, an educational technologist and a subject sophisticates (i.e. a student who has successfully completed the course)"[4] while the one-to-one evaluation consisted of ten fourth year undergraduate students who were studying Multimedia Technology Systems (2003, p. 11).

In regards to program control mode, subjects suggested that the instructional guidance at the bottom of the screen should be reduced to give more screen space to the content.[4] Hyperlinks were colored by importance to persuade or dissuade students' choices, however, the subjects did not like that the gray links(signifying content deemed currently unsuitable for the user) were not click-able(2003, p. 11-12).

Following the feedback, the evaluators improved the system and started a new phase. For the third phase, ten more fourth year undergraduate Multimedia Technology Systems students were selected based on their score making sure it fell within one standard deviation of the mean. The selection of the subjects was in line to represent the actual student population.

Following the third phase, the feedback showed that the majority of the students were satisfied with the initial adaptation based on their cognitive style and that they found useful the ability to change the initial stage through the student model. They also indicated as very important, having different instructional modes in order to accommodate their individual needs. Finally, they were satisfied with the ability to have complete control over how the system served them content.[4]

Triantafillou's iterative design of a hypermedia system takes steps towards creating a successful adaptive model. Two modes of control exist to the user allowing them more or less freedom to peruse the content of the course. Both users who used the system with program control and learner control successfully improved their pretest performance to their post-test performance. More research is necessary to see if students utilizing program control can achieve levels of improvement on par or better than students utilizing learner control.[4]

Katholieke Universiteit[edit]

Conducted by Denise Pilar da Silva, Rafaël Van Durm, Erik Duval, Henk Olivié of Katholieke Universiteit Leuven, Belgium. Similar to the Aristotle University study, a prototype adaptive hypermedia system was created. This particular system uses three frames within the single browser window to display the navigation tree, relevant content links, and the content itself. It also utilized a link hiding feature to remove links to content already viewed, however, this was ultimately received as confusing.

Some important conclusions from this study include redesigning navigation trees to show more localized navigation and less overall hierarchy in order to reduce cognitive overhead. Additionally they concluded it is worthwhile to redesign the measures of difficulty and concept coverage so that it is in direct relation to a user's knowledge model. However, the main problem encountered is how to properly define relationships between documents and difficulty ratings and thresholds in proper relation to the user and content.[6] Also importantly, it is noted that proper knowledge of the content in the course is important when deciding relationships between content. In other words, designing an adaptive educational course might be best done on an individual basis rather than creating an all encompassing format for all courses of varying areas of focus.

See also[edit]

External links[edit]


  1. ^ Gerherd Weber; Peter Brusilovsky (2003). "ELM-ART: An adaptive versatile system for Web-based instruction". International Journal of Artificial Intelligence in Education 13 (2–4): 159–172. 
  2. ^ Peter Brusilovsky; John Eklund (1998). "A study of user-model based link annotation in educational hypermedia". Journal of Universal Computer Science 4 (4): 429–448. 
  3. ^ Paul De Bra; Licia Calvi (1998). "AHA! An open Adaptive Hypermedia Architecture". The New Review of Hypermedia and Multimedia 4: 115–139. 
  4. ^ a b c d e f g h Triantafillou, E, Pomportsis, A, & Demetriadis, S. (2003). The design and the formative evaluation of an adaptive educational system based on cognitive styles. Computers & Education , 41(87-103), Retrieved from doi:10.1016/S0360-1315(03)00031-9
  5. ^ a b c Peter Brusilovsky (2003). "Developing Adaptive Educational Hypermedia Systems: From Design Models to Authoring Tools". Authoring Tools for Advanced Technology Learning Environments: Toward cost-effective adaptive, interactive, and intelligent educational software. Kluwer. ISBN 978-1-4020-1772-8. 
  6. ^ Silva, S.P., Durm, R.V., Duval, E., & Olivié, H. (1998). Concepts and documents for adaptive educational hypermedia: a model and a prototype. 2nd Workshop on Adaptive Hypertext and Hypermedia, Retrieved from