A deterministic crowding evolutionary algorithm to form learning teams in a collaborative learning context

► We address the problem of forming collaborative learning teams from a students' class. ► Grouping criterion: well-balanced teams according to the roles of their members. ► To solve the problem, we propose a knowledge-based evolutionary algorithm. ► The balance of the teams is evaluated based on knowledge of the students' roles. ► The approach has reached excellent results on ten different data sets.