Ensemble Machine Learning Model Improves Prediction Accuracy for Academic Performance: A Comparative Study
Published in Bioengineering & Biotechnology
Overfitting is a common issue in machine learning models, where the model only predicts training data, but can be resolved using ensemble techniques like bagging or boosting. Bagging involves predicting accuracy through majority voting of multiple models running in parallel, while boosting uses a weak sequential model and passes the previously overfitted samples to another model to generate a more robust prediction. This research tries to fill the research gap between the selection of the best model for classification problem after comparing accuracy score of default and boosting algorithm. The multiple linear regression summary statistics indicate the dependent variables were significant, so prediction accuracy needs to be further tested using an ensemble machine learning model. The logistic regression model scored the least in both default (72%) and cross-validation (84%) accuracy than the random forest
I'm Yagyanath Rimal, a university teacher specializing in computer science. In addition to teaching C, C++, Java, and web applications at the bachelor's level, I also teach Research Methodology, Data Analytics, and Machine Learning to master's degree students. Currently, I am focusing on research in data analytics and machine learning. I've recently embarked on my journey in academia and am also interested in international collaborations and postdoctoral opportunities.