Group formation is an important topic in Computer Supported Collaborative Learning (CSCL) because that has implications in the group performance. In this paper, we propose a genetic algorithm for automatic generation of groups considering learning styles of your members. The group formation with genetic algorithm is a permutative problem, for this reason, genetic operators were designed. We use historical data about performance of groups and we create association rules which are used in the fitness function. The algorithm proposed was analyzed with different size of groups given for the teacher. Through the experimentation we can see what kind of configuration tends to be more appropriate.

For personal information agents, user profiles have to represent user interests and preferences in order to satisfy long-term information needs. An implicit assumption in user-profiling is the existence of persistent interests which, however, might suffer some changes over time. Each time the interests of a user change, his profile becomes inaccurate and the predictive quality decreases. Adaptation of user profiles is, therefore, an essential requirement for personal agents that need to be capable of adjusting their behavior quickly in order to shorten the period of reduced predictive quality. In this paper, a user-profiling technique named WebProfiler, which learns a hierarchical representation of user interests using conceptual clustering, is augmented with an adaptation strategy based on relevance feedback and time-based forgetting in order to deal with drifting interests. We empirically evaluate the performance of this strategy by analyzing its behavior on multiple scenarios of interest drifts and shifts.

Computer Supported Collaborative Learning (CSCL) systems have recognized advantages. However, using these systems does not guarantee an effective collaborative learning. Success or failure of the learning experience depends on the collaborative skills the students show in the group. This work presents a multiagent model applied to CSCL environment, which aims both at recognizing conflicts occurring in group dynamics and at providing personalized training of collaborative skills demonstrated by group members. Conflicts are recognized by applying the Interaction Process Analysis method. Personalization is achieved through Bayesian networks that consider students' collaborative characteristics to elucidate the most suitable training strategy. The model was implemented in a distance learning environment and showed great efficacy.

An effective collaboration in learning environments involves a set of skills that students must learn and cultivate. Detecting the contexts in which students apply these skills allows learning environments to offer personalized assistance during the learning process. In this paper a Collaborative Profile, as part of a student model, is introduced. This profile captures collaborative skills of a student working in a group. To build the Collaborative Profile a Web Usage Mining Approach is applied and collaborative behaviour patterns are detected automatically, identifying contexts in which student's skills are evident. This profile is designed to offer additional information to improve personalized assistance in collaborative learning environment.

The motivation behind personal information agents resides in the enormous amount of information available on the Web, which has created a pressing need for effective personalized techniques. In order to assists Web search these agents rely on user profiles modeling information preferences, interests and habits that help to contextualize user queries. In communities of people with similar interests, collaboration among agents fosters knowledge sharing and, consequently, potentially improves the results of individual agents by taking advantage of the knowledge acquired by other agents. In this paper, we propose an agent-based recommender system for supporting collaborative Web search in groups of users with partial similarity of interests. Empirical evaluation showed that the interaction among personal agents increases the performance of the overall recommender system, demonstrating the potential of the approach to reduce the burden of finding information on the Web.