Intelligent Agent to Form Heterogeneous Group Based On Personality Traits with Genetic Algorithm

Group formation is the first step and has an important effect on collaborative learning based on CSCL (Computer Support for Collaborative Learning). Heterogeneous groups have been tested to increase interaction in collaborative learning. This research develops intelligent agents based on genetic algorithms to form heterogeneous groups. The level of group heterogeneity is built based on the student's personality traits. Genetic algorithms are used as a method of intelligence to make diversity within a group. Intelligent agent performance is measured based on heterogeneity, and collaboration performance. Intelligent agent testing produces good scores for heterogeneity and collaboration performance. It can be concluded that intelligent agents are successful in forming heterogeneous groups. Heterogeneity based on students is proven to improve the quality of collaborative learning.

I. Introduction

Collaborative Learning (CL) is defined as a group learning method to achieve the common goals of students [1]. It can take the form of dialogue, negotiation, and arguments to solve the problems they have. The implementation of this learning has three model dimensions. The dimensions of the learning model include: (1) a pair or more people working together; (2) credible learning material (namely learning material); and (3) collaborative interaction methods between group members [2].