A method for improving learning effect by using a prediction model and an optimization algorithm

Abstract

In this paper we propose a method to determine the optimal combination of learners to improve learning effect through approaches of system engineering. In the proposed method, a neural network model is first applied for predicting the learning results of learners. This model has input values which are potential abilities and personalities of a pair of learners and has one output value which is the correctness rate of a learner of the pair for learning problems. Secondly, the optimal combination of the learners is determined by using a genetic algorithm and the prediction results. This proposed method can realize the improvement of learning effect by a simple process. © 2008 SICE.