Prediction of Student Decision-making Behaviour based on Machine Learning Algorithms

Abstract

The study of student decision-making behaviour holds immense importance in the realm of education as it aids educators in comprehending students' learning patterns, preferences, and possible scholastic achievements. Nevertheless, the majority of existing research on behavioural decision-making primarily examines the influence of decision-making on students using questionnaires, disregarding the significance of technological methods in forecasting decisions. This study presents a technique for forecasting student decision-making patterns through the utilisation of machine learning techniques. This work utilises many machine learning techniques, including k-nearest neighbour, decision trees, support vector machines, and linear regression, to develop prediction models. These models are created by analysing students' historical data and attributes. The experimental findings substantiate that the machine learning algorithms put forth have the capability to accurately forecast student decision-making behaviour with a significant level of precision. The prediction outcomes can serve as significant guidance for educators to design customised teaching approaches, thus enhancing students' academic achievements and contentment.