Student's Performance Prediction Using Decision Tree Regressor

A particular form of artificial intelligence termed machine learning, or ML, has multiple uses in the education sector. Organizations are using ML solutions more frequently to improve customer experiences, increase ROI, and gain a competitive advantage. Additionally, it predicts that this cutting-edge approach will be widely adopted by the education sector. Therefore, it is accurate to say that ML will significantly influence the future of the education industry as we transition to the new educational system. Machine learning in education can appear to have become just another trendy term designed to compel business owners to spend money on self-indulgent innovation. We focus on predicting student performance as our area of machine learning in education. As personalization and customized services are becoming more popular, personal assistance for students may be thought about in the education sector. We may predict a student's performance based on their past behavior, health, family background, and geographic dispersion and we generally analyze and report that performance in an effort to make it better. Based on extremely high Dimensional data, we tend to discard data frames that have little bearing on student performance in favor of building models that provide the maximum degree of accuracy. These statistics look at secondary school student achievement at two Portuguese schools and retrieved from https://archive.ics.uci.edu/ml/datasets/student+performance. In this paper, we examine many aspects that impact students' performance and pinpoint the most crucial features among them, in order to construct a machine learning model with the best level of accuracy. We have used four different algorithms and compared their accuracy. It includes Linear Regression, Lasso Regression, Decision Tree Regressor and Random Forest Regressor. The accuracy of the Decision Tree Regressor is highest with a value of 99.94%. All models are evaluated on the basis of RMS Error.