International Forum of Educational Technology & Society

Formal Discussion Initiation

Adaptive Patterns in Authoring of Educational Adaptive Hypermedia

(Participation in the discussion requires free membership of the forum.)

Time schedule:
Discussion: 14-23 July 2003
Summing-up: 24-25 July 2003

Alexandra Cristea
Technical University Eindhoven, The Netherlands


Problem issues

One of the main problems with e-learning environments is their lack of personalization (or adaptivity). They cannot offer customisation for the student; they can only offer identical contents to all the consumers.

Educational Adaptive Hypermedia (EAH) environments should be the answer to this problem. Adaptive Hypermedia is a relatively new field (starting around the early 1990s), but it already has reached some maturity and already features some good applications.

Imagine, for instance, that, instead of showing all students an example of a painting by Rembrandt and then show them the description of the chiaroscuro technique, one could write a rule:

IF student. chiaroscuroTechnique.motivation=’non-existent’ THEN show.paintingRembrandt;

So, show the painting only to students without motivation in studying the chiaroscuro technique (with the hope that seeing such a great painting will motivate them), and skip this step if the motivation is already there (give direct information for motivated students). This very simple rule already offers some customisation, dependent on a student model variable called ‘motivation’.

At the Eindhoven University of Technology, for instance, a few courses are given already for a couple of years on-line, in adaptive hypermedia environments (featuring student models and automatic adaptation to student’s computed ‘knowledge level’: e.g.,

However, such EAH environments are not yet as spread as we would have expected, and not many educators even know about the opportunities offered by such automatic personalization, let alone make use of it.


Our opinion about those issues

The major drawback that we have found is that, at present, there are no good authoring tools for educational adaptive hypermedia, and creating EAHs from scratch without such tools becomes actually a programmer’s job – and not that of an author (e.g., teacher).

Therefore, we embarked on the quest of designing appropriate authoring tools for EAHs.

Please note that such authoring tools as we envision do not try to compete with commercial tools such as Blackboard, WebCT, etc. What we want is to complement these tools with what they lack: the possibility of expressing the automatic personalization elements for the student.

Please also note that such authoring is normally more difficult than authoring of a sequential course: alternatives have to be designed, linked and annotated (i.e., the static descriptions). The rules of when and how to take the alternative routes have to be specified (i.e., the dynamics description).

In order to make authoring efficient, such an authoring system should propose to the author a lot of methods of reuse, both of contents (static description) as well as of contents dynamics.

The static contents description reuse has been discussed and researched more thoroughly, and it can be based on the learning object model (LOM), on the emerging SCORM standard, as well as on work on  hypermedia patterns as developed by the team of Franca Garzotto ( and on our own work on course domain structuring ( [4].

Therefore, our main focus is to research and work on the re-usage of the dynamics, i.e., on the re-usage of adaptation features. Please note that the IMS ADL simple sequencing protocol, although a step in the right direction, cannot provide all the features of adaptation that adaptive hypermedia can offer.


Suggested solutions

We are running a European Community project, ADAPT, whose main goal is to extract adaptive patterns of educational adaptive hypermedia, and to use these in authoring:

Our purpose is to find these more general patterns of adaptive behaviour, and then make them reusable via an authoring environment that offers these patterns, or even groups of patterns (adaptation strategies), as we shall see. To go back to the adaptation example above, the IF-THEN rule described there cannot be reused: it is too specific, binding two specific instances, paintingRembrandt and chiaroscuroTechnique. Moreover, if we would want to make a similar connection between Picasso and the collage technique, we would have to actually write a new rule:

IF student. collageTechnique.motivation=’non-existent’ THEN show.paintingPicasso;


Obviously, the two rules have much in common, and it is natural to expect that an author would want to use a similar structure for the Picasso case, which s/he can reapply instead of writing a new rule. To make this rule reusable, we should be allowed to write higher (schema) level constructs such as:

IF student. Technique.motivation=’non-existent’ THEN show.painting;

Show. Technique;

The advantage of the rule above is not only that it summarizes the rules for Picasso’s and Rembrandt’s paintings, but that it can connect any other painter and the technique s/he invented. This is just a very simple example how generalization of the content structure can lead to reusable components, even for adaptation rules. However, we can go even further than that, and notice that the two higher level constructs, technique and painting, can be seen as representing some theory and its corresponding example, and that they could be both belonging to a concept that binds the theoretical part with its practical part. As we can then define a relation that connects this theoretical part of the concept with its practical part, by marking the example as a specialization of the theory (and, vice versa, the theory as a generalization of the example), we can replace the previous rule with a higher-level language rule, as follows:

IF student.Concept.motivation=’non-existent’ THEN specialize;


From these kinds of considerations the first constructs of an adaptation language emerged, as described more systematically in [1]:

The idea behind it was, as mentioned above, to let the author of adaptive educational hypermedia work on a higher semantic level, instead of struggling with the ‘assembly language of adaptation’. Furthermore, these patterns should represent the first level of reusable elements of adaptation, as shown in the examples above.

However, reusability can go further than that. Even this adaptation language, we are aware, might still be difficult to handle for some authors (teachers). So, as mentioned in the paper above ([1]), reuse should be strived at even at the level of adaptation strategies (that correspond to cognitive/ learning strategies).  In this paper we show how the four cognitive styles identified by Kolb (converger, diverge, assimilator, accomodator) can be written in adaptation language (as well as in adaptation assembly language) and transformed into adaptation strategies, ready to be reused. Below (table 1) is an extract example from the paper with the implementation for the cognitive style converger, using the specialize and generalize adaptation language constructs, as described previously, but also some other special construct, such as enough. The latter is a more relaxed way to specify preconditions, without precisely defining them. In this case, enough(result) means, e.g., having a passing average for at least two of the tests, but other definitions are also possible.

medium_increase() : generate adaptive presentation with (obviously) increasing difficulty

1. Explanation: Convergers are abstract and active; they like to feel in control; start with course for intermediates at medium adaptivity level, repeat for a number of times:

  • evaluate state of learner and start increasing difficulty & decreasing adaptivity  level if result=good
  • evaluate state of learner and start decreasing level if result=bad

2. Translation at medium level: (ENOUGH shows here that the result is above an average result)

AdaptLevel= 5; N=AskUser(); # this is to let user feel and be in control; levels: (1=min  to 10=max)

FOR <I=1..N> DO

{   SPECIALIZE (ENOUGH(Result));  IF (AdaptLevel>1) AdaptLevel--;

     GENERALIZE (NOT(ENOUGH(Result))); IF (AdaptLevel<5) AdaptLevel++;

} # Note that adaptation level is not allowed to increase too much

3. Translation at low level: (the average can be implemented but takes more space)

DiffLevel = 3; AdaptLevel= 5; # note that here there is no predefined number of repetitions

IF <ACTION> THEN # Note that above we don’t need the action of the user for triggering;

{    IF (Result1 +Result2)/2>5 AND DiffLevel<10 THEN #  Note that ‘enough’ and specialize

         { DiffLevel++;  IF (AdaptLevel>1) AdaptLevel--;}  #  must be redefined each time

     IF  (Result1 +Result2)/2<5 AND DiffLevel>1 THEN {DiffLevel--; IF (AdaptLevel<5) AdaptLevel++;}


Table 1. Adaptation strategy for cognitive style: converger (abstract, active)


In discussions resulting at the User Modelling conference, other cognitive styles that should be implemented were suggested, such as field dependence versus field independence.


Questions for the members to respond to

  1. What other cognitive/learning strategies would be appropriate to experiment with for e-learning in particular? (i.e., to transform them into adaptation strategies). The main target we are envisioning are university students, but extensions towards other categories of pupils are also important, especially in the context of life-long learning.
  2. Are the patterns described so far enough for describing the cognitive/ learning strategy you would like to implement?
  3. What other patterns would be necessary, and for what cognitive/ learning strategies?
  4. Would you opt for reusing directly adaptation strategies, or would you prefer to write your own?
  5. If you would write your own adaptation strategies, what would you need in the adaptation language?
  6. What criteria/ constraints should the adaptation language fulfil?


Papers produced by this research

  1. Cristea, A.I., and Calvi, L. The three Layers of Adaptation Granularity. UM’03. User Modelling - June 2003. Springer.
  2. Cristea, A.I. Automatic Authoring in the LAOS AHS Authoring Model. HT’03. Hypertext workshop on AH. ACM (in press).
  3. Cristea, A.I., and Kinshuk. Considerations on LAOS, LAG and their Integration in MOT. ED-MEDIA’03.
  4. Cristea, A., De Mooij, A. LAOS: Layered WWW AHS Authoring Model and its corresponding Algebraic Operators. In Proceedings of WWW’03, Alternate Education track. (Budapest, Hungary 20-24 May 2003). ACM.
  5. Cristea, A., De Mooij, A. Designer Adaptation in Adaptive Hypermedia. In Proceedings of ITCC’03 (Las Vegas, US 28-30 April 2003) IEEE Computer Society.
  6. Cristea, A.I., and De Bra, P. Towards Adaptable and Adaptive ODL Environments. In Proceedings of AACE E-Learn’02 (Montreal, Canada, October 2002), 232-239.
  7. Cristea, A.I., Okamoto, T., and Kayama, M. Considerations for Building a Common Platform for Cooperative & Collaborative Authoring Environments. In Proceedings of AACE E-Learn’02 (Montreal, Canada, October 2002), 224-231.
  8. Calvi, L., and Cristea, A.I. Towards Generic Adaptive Systems Analysis of a Case Study. In Proceedings of AH’02 (Malaga, Spain, May 2002) Adaptive Hypermedia and Adaptive Web-Based Systems, LNCS 2347, Springer, 79-89.
  9. Cristea, A.I., and Aroyo, L. Adaptive Authoring of Adaptive Educational Hypermedia, In Proceedings of AH 2002, Adaptive Hypermedia and Adaptive Web-Based Systems, LNCS 2347, Springer, 122-132.
  10. Cristea, A., De Mooij, A. Adaptive Course Authoring: MOT, My Online Teacher. In Proceedings of ICT-2003, IEEE LTTF International Conference on Telecommunications, "Telecommunications + Education" Workshop (Feb 23 - March 1, 2003TahitiIsland in Papetee - French Polynesia).



About moderator

Alexandra Cristea is Assistant Professor at the Technical University Eindhoven, The Netherlands.

Copyright message
Copyright by the International Forum of Educational Technology & Society (IFETS). The authors and the forum jointly retain the copyright of the articles. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear the full citation on the first page. Copyrights for components of this work owned by others than IFETS must be honoured. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from the editors at

Back to discussions schedule Page Back to Main Page