Generic adaptation framework for unifying adaptive web-based systems

E. Knutov

Research output: ThesisPhd Thesis 1 (Research TU/e / Graduation TU/e)

461 Downloads (Pure)

Abstract

The Generic Adaptation Framework (GAF) research project first and foremost creates a common formal framework for describing current and future adaptive hypermedia (AHS) and adaptive webbased systems in general. It provides a commonly agreed upon taxonomy and a reference model that encompasses the most general architectures of the present and future, including conventional AHS, and different types of personalization-enabling systems and applications such as recommender systems (RS) personalized web search, semantic web enabled applications used in personalized information delivery, adaptive e-Learning applications and many more. At the same time GAF is trying to bring together two (seemingly not intersecting) views on the adaptation: a classical pre-authored type, with conventional domain and overlay user models and data-driven adaptation which includes a set of data mining, machine learning and information retrieval tools. To bring these research fields together we conducted a number GAF compliance studies including RS, AHS, and other applications combining adaptation, recommendation and search. We also performed a number of real systems’ case-studies to prove the point and perform a detailed analysis and evaluation of the framework. Secondly it introduces a number of new ideas in the field of AH, such as the Generic Adaptation Process (GAP) which aligns with a layered (data-oriented) architecture and serves as a reference adaptation process. This also helps to understand the compliance features mentioned earlier. Besides that GAF deals with important and novel aspects of adaptation enabling and leveraging technologies such as provenance and versioning. The existence of such a reference basis should stimulate AHS research and enable researchers to demonstrate ideas for new adaptation methods much more quickly than if they had to start from scratch. GAF will thus help bootstrap any adaptive web-based system research, design, analysis and evaluation.
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Department of Mathematics and Computer Science
Supervisors/Advisors
  • De Bra, Paul M.E., Promotor
  • Pechenizkiy, Mykola, Copromotor
Award date9 May 2012
Place of PublicationEindhoven
Publisher
Print ISBNs978-90-386-3131-8
DOIs
Publication statusPublished - 2012

Fingerprint Dive into the research topics of 'Generic adaptation framework for unifying adaptive web-based systems'. Together they form a unique fingerprint.

Cite this