A Software Engineering Model for the Development of Adaptation Rules and its Application in a Hinting Adaptive E-learning System

Pedro J. Muñoz-Merino, Carlos Delgado Kloos, Mario Muñoz-Organero, Abelardo Pardo

The number of information systems using adaptation rules is increasing quickly. These systems are usually focused on implement nice and complex functionality for adaptation of contents, links or presentation, so software engineering methodologies for the description of rules are required. In addition, the distributed service oriented Internet philosophy presents the challenge of combining different rules from independent Internet sources. Moreover, easy authoring, rule reuse and collaborative design should be enabled. This paper presents the AR (Adaptation Rules) model, a new software engineering model for the description of rules for adaptation. These rules can be composed as a set of smaller atomic, reusable, parametric, interchangeable and interoperable rules, with clear restrictions in their combinations. Our model enables the distribution of rules as well as rule reuse and collaboration among rule creators. We illustrate our approach with the application of this model to a hinting adaptive e-learning system that generates exercises with hints, which can be adapted based on defined rules. Advantages of the AR model are confirmed with an evaluation that has been done with teachers and learning analytics experts for adaptive e-learning.