Hamid Sadeghi - Department of Computer Engineering, Hashtgerd Branch, Islamic Azad University, Hashtgerd, Iran
Mina Etehadi Abari - Department of Computer Engineering, University of Isfahan, Isfahan, Iran

Nobody can cast doubt on the fact that the advent of computer-supported collaborative learning (CSCL) hasmade a great revolution in educational systems. The first phase in CSCL is creating different learning groups withcompatible members. It is a key factor because having a prosperous collaboration depend on having compatibleteam-mates to do their tasks collaboratively. Although, a great number of studies in the literature reported severalmethods for assigning students to fitting groups all of them are bereft of offering a completely independent andintelligent system. This contribution suggests a neural-network grouping method in order to enhance speed,simplification, and correctness of group composition. Extensive experiments and evaluation illustrate the success ofour method and students are more satisfied and seize high knowledge levels when they are grouped via the proposedmethod.

Computer-supported collaborative learning; CSCL; Neural network; MLP; Group composition