Hard Voting Classifier Based Students Performance Prediction Using Machine Learning

Abstract

Ensemble learning is an imperative feature of machine learning. It helps in improving the prediction accuracy. The research paper uses the ensemble techniques predict the performance of students. Holistic development of students is essential at every stage of life. Predicting the performance of student to avoid dropout ratio becomes important to improve the outcome of student’s performance in field of education. It includes the various factors that have an impact on students’ performance. The dataset used is from Kaggle Higher education students’ performance evaluation. The performance of individual classifier is evaluated that is Decision Tree classifier, Gradient Boosting model and Random Forest which is further combined. The Voting Classifier is applied to advance the achievement of the collaborative model which gives the accuracy of 86%.