A Random Forest Algorithm for Predicting Computer Programming Skill Associated with Learning Styles

The complexity of programming concepts such variables, loops, arrays, and functions contribute roadblocks for students learning to code. Predicting computer programming skills using machine learning is commonplace. It enables early identification of student at risk of programming failure and prompt implementation of successful early intervention strategies. Artificial intelligence relies to acquire information and rules from complicated data in order to foresee outcomes and patterns in behavior. In contrast to statistical approaches, machine learning seeks to improve prediction performance by making more accurate forecasts. Hence, we set out to investigate, using machine learning techniques, partially the Random Forest Algorithm’s (RFA) to predict vocational high school students’ proficiency in computer programming. The outcomes indicated a pass prediction of 90.23 percent, a failure prediction of 55 percent, an overall accuracy of 88.79 percent, and a total performance indicator of 91 percent across all classification cutoffs.