Decision tree algorithm for predicting student performance based on psychological tests
Limbong, San A.
Decision tree algorithm for predicting student performance based on psychological tests - Vol.5(2), May-Aug - New Delhi Enriched Publications 2024 - 1-12p.
It is essential to consider the psychological aspect of selecting new students to determine the success of
prospective students. In this paper, we propose an approach to predict student performance based on their
psychological test scores using the Decision Tree algorithm. The dataset used in this study was taken from
the student admission process at the Institut Teknologi Del.
The admission dataset contains the scores of psychological tests and the Grade Point Average (GPA) of
classes 2019, 2020, and 2021. Each class has its own attribute set. Therefore, we came up with two
approaches. The first approach was to use as many records as possible, and the opposite of the second was
to utilize more features.
Our results showed that the first approach was slightly better. The MAE value was 0.3654 to 0.4568.
Moreover, none of the psychological test attributes strongly correlate to GPA and hence do not guarantee
student performance.
Computer Engineering
Decision tree algorithm for predicting student performance based on psychological tests - Vol.5(2), May-Aug - New Delhi Enriched Publications 2024 - 1-12p.
It is essential to consider the psychological aspect of selecting new students to determine the success of
prospective students. In this paper, we propose an approach to predict student performance based on their
psychological test scores using the Decision Tree algorithm. The dataset used in this study was taken from
the student admission process at the Institut Teknologi Del.
The admission dataset contains the scores of psychological tests and the Grade Point Average (GPA) of
classes 2019, 2020, and 2021. Each class has its own attribute set. Therefore, we came up with two
approaches. The first approach was to use as many records as possible, and the opposite of the second was
to utilize more features.
Our results showed that the first approach was slightly better. The MAE value was 0.3654 to 0.4568.
Moreover, none of the psychological test attributes strongly correlate to GPA and hence do not guarantee
student performance.
Computer Engineering