Engineering Adaptive Concept-Based Systems for the Web

Abstract

In recent years we have witnessed a growing interest in adaptation and personalization in numerous application domains including business, education, and so forth. Applications that offer large bodies of information have in the Web era turned into systems with a significantly different nature than two decades ago. Think of the typical book catalog database from 20 years ago and the Web site of a book seller nowadays. A characteristic aspect of the restyling is the attention paid to the individual user. Technology has evolved and now allows application designers to include adaptation and personalization in the applications. This is especially important in the field of e-commerce, where users (customers) expect personalized services similar to those they receive in a conventional store. Typical e-commerce systems employ large bodies of information. In the case of a bookstore Web site, the designer defines an appropriate structure for the collection of books with all relevant properties. Generally, the design uses structures of concepts, where the concepts represent the actual information objects. The adaptation engineering is later performed on the level of these (abstract) concepts. We refer to these applications as concept-based systems. Adaptive concept-based systems are especially accepted in areas where the main goal is to tailor large amounts of information to the individual preference and knowledge state of the user. Besides electronic commerce, other examples include online museums (the visitor wandering through the collection on an individual basis) and e-learning applications (the student being involved with learning material in a way that the teacher thinks fits the student’s situation).