Student Success Performance Prediction Using Artificial Intelligence Approach

Prediction of student success is one of the numerous Artificial Intelligence (AI) applications that assist instructors in identifying students that need individualized support. Due to this factor, student behavior in institute may affect the learning results less scores in their academic performances. Intelligent algorithms have utilized this task considered for several factors for making accurate decisions. However, the traditional models have generated decisions are inefficient because of lacking in trust and explainability. The dataset collected with 33 attributes of personal details of students which include academic as well as student’s behavior associated inside and outside campus of institute. However, finding of behavioral factors that affects the APS is essential to avoid and improve the student attitude and behavior in a better manner using Machine Learning (ML) methods but determines less accuracy. The research revealed that these models deemed certain sensitive features to be significant raising concerns about trust and fairness in their use. Hence, this paper have implement hyperparameter tuning of ML using Randomized Search optimizer has efficient in providing better balanced accuracy in determining the high classification prediction in ML method which determine the interest in academic as well as attitude of students in the university.