Case-based reasoning approach to Adaptive Web-based Educational Systems Paulo Alves Instituto Politécnico de Bragança palves@ipb.pt Luís Amaral Universidade do Minho amaral@dsi.uminho.pt José Pires Instituto Politécnico de Bragança adriano@ipb.pt Abstract Virtual learning environments systems are based on the classroom paradigm, in which the knowledge is transmitted for all the students in the same way. To enhance e-learning, adapting the contents to the needs of each student is essential, and a more personalized learning support is required. The adoption of pedagogical agents and new artificial intelligence methodologies can response to the needs of individual students and provide a more effective collaboration in virtual learning environments. The learning experience of each student can be adapted to others students with the same characteristics. The adaptation of past cases to solve new problems is one of the features of case-based reasoning methodology, which can provide an effective knowledge transmission, based on learning activities. In this paper, we present a case-based reasoning approach to Adaptive Web-based Educational Systems using fuzzy logic to adapt e-learning contents and contexts according to the student learning style and individual needs. 1. Introduction The Bologna Process, has the main goal to establish a European Higher Education Area by 2010. The institutions are faced with new challenges in the structural change on curricula and the adoption of innovative teaching and learning processes. To face this challenge, the curricula design must be based on learning outcomes and less in the classical teaching model, integrating e-learning technologies and new pedagogical models. The pedagogical model is centered on the student and the lecturer has the mission to support the learning process in the contact hours as so at distance using information and communication technologies. To improve the learning process and facilitate the student support, we propose the adoption of agents in Adaptive Web-based Educational Systems, with the mission to support the student in their learning activities, coaching, advertising for difficulties, and adapting contents and contexts according to each student learning path. 2. Case-Based Reasoning Case-based reasoning (CBR) is one of the major reasoning paradigms in artificial intelligence, with applications in several research areas like healthcare, education, business, legal reasoning, and manufacturing. Kolodner [1] defined CBR as adapting old solutions to meet new demands. CBR has the advantage of the low initial training of the system, compared with other expert systems like rule-based reasoning [2]. In CBR the relation of the problems with their solutions is obtained from experiences and the system can start operate with few stored cases. The problem-solving life cycle in a CBR systems described by Aamodt and Plaza [3] consists essentially of the following four parts: retrieving similar cases experienced in the past, reusing the cases copying or integrating the solutions from the cases retrieved, revising or adapting the solution(s) and retaining the new validated solution The use of CBR in education has several advantages like supporting the lecturers in the design of learning activities or to improve the student’s knowledge. We use this approach in educational agents, using the past cases to solve new problems in the learning process. This agent has the characteristic of an advisor that alerts the student for all the events, coaching the student and coordinating the collaborative activities. The agent has also the mission to support the student in the agenda management, assessment, project