User Modeling 2007
Towards User Modelling and Adaptive Systems for All (TUMAS-A)

Submission instructions

The following experts raise some open problems in the field and post challenging questions. Participants of the workshop should select some of these open questions and try to answer them from their experience. Format and procedure to submit the papers are provided at the end.

Adaptation Issues in TUMAS-A 2007

Expert 1. Adaptation issues

There are different types of adaptation issues that affect the implementation of accessible learning environments for ALL.

First, we are assuming a LLL paradigm that guarantees accessibility from any device, by anyone, at any location. In this paradigm multiple scenarios will have to be encompassed, i.e. users with different abilities and preferences making use of learning services through network enabled devices with diverse computing and interaction capabilities. Within this context adaptation features are supposed to provide the required support to the information exchange between services, devices, user's preferences and context.

Second, adaptive learning environments are the result of the evolution of three related areas: Intelligent Tutoring Systems, Adaptive Hypermedia Systems and Computer Supported Collaborative Learning systems. Moreover, adaptation is not an idea that can be plugged in a learning environment, but a process that influences the full life cycle of learning. If we analyze current LMS, working on learning environments is a complex process of four interrelated steps: design of the learning experience (based on objectives and learning activities), administration (i.e., management of all data including users' roles, access rights and services configuration), usage (i.e., actual use of designed activities on the learning environment within the class context) and auditing (i.e., authors get reports on the actual use of course design, namely descriptions on how users have performed on learning activities, in order to adjust course design).

Third, to provide adaptive learning environments which consider such variety of services, devices, user's preferences and context, diverse knowledge-based methods and machine learning techniques should be applied. Furthermore, these techniques have to be combined with educational standards (IMS-CP, IMS-LIP, IMS-QTI, IEEE-LOM/IMS-MD) to control the corresponding feedbacks between design decisions and runtime interactions.

The above three related issues give rise to a wide variety of open questions as far as adaptation mechanisms are concerned. Some of them are as follows:

  • What types of characteristics are to be considered in the models to support required adaptations?
  • What type of user, usage and environment data are needed in order to construct accessible and adaptive user models?
  • To what extent available standards cover a wide feature range to support the required adaptations?
  • Which are the more appropriate techniques to construct adaptive models from interaction data?
  • How to accommodate the presence of multiple devices and assistive add-on technologies in the adaptation process?
  • What types of architectures are more suitable to support adaptive and accessible resource delivery?

Expert 2 - eAccessibility and eLearning

As business and society become more and more dependent on information and communication technologies as well as embedded software systems, the impact of the Digital Divide caused by differences in accessibility to the new technologies could grow by widening differences in educational chances, job market situation, interaction with public administration and government, and quality of life.

Lifelong Learning (LLL) is nowadays introduced in the form of eLearning, which is intended to be available in any environment and with any device. This is a challenge that affects, not only people with special needs, but all users. Within LLL, the term user includes both students and teaching staff. These constraints demand that interfaces, content and presentation need to be adapted to:

  • different devices, e.g., desktop, PDA, Smart Phone, etc.
  • different contexts of use, e.g., home, workplace, commuting, etc.
  • different types of content and content generators (i.e., LMS)
  • different learning objectives and activities and
  • different user characteristics and assistive technologies

To combine all these factors, there are several technologies that come into play, like Web Services, Semantic Web and Content Metadata. Whilst the industry has gone a long way in implementing some of these technologies in the area of Device Modelling for mobile devices (via CC/PP and UAProf), a lot of work needs to be done in other areas, where research or implementation is in its infancy.

This brief analysis leads to several open questions:

  • How can existing work in the area of user modelling can be extended to reflect the needs of this ubiquitous learning environments? (See, e.g., AccLIP standard.)
  • How can existing standards in the area of interface modelling like URC, or research in the area of fluid computing, contribute to the adaptation process?
  • Which services would be critical to combine user and device profiles to facilitate content adaptation?
  • Do educational standards (see previous set of questions) need to be extended/modified to cope with these new demands?
  • Are Semantic Web technologies adequate to model all the factors outlined earlier?
  • Are truly adaptive interfaces a feasible objective, or are the needs of the target population better covered with adaptations based upon user stereotypes?

Format for preparing the papers

The format for the workshop papers should be the standard Springer format used for the main conference BUT with narrower margins:
TOP and BOTTOM - 3cm;
LEFT and RIGHT - 3.5cm.

The length of the papers will be 6 pages (extended) with direct answers to some of the challenging open questions. The authors are encouraged to answer only those questions that have been previously address in their research works.

Each paper will be pair reviewed by 3 experts from the Program Committee, who will provided valuable feedback for the work.

Workshop papers will be published in the workshop proceedings by the UM2007 Organizing Committee and also will be available on the workshop Web site.

Submissions should be in Open Document Format (ODT file), although RTF format is also accepted.

Files should be sent to Olga C. Santos and Jesus G. Boticario by email.