Performance Evaluation of Different Machine Learning Algorithms on Student Dataset Clustered by K-means Algorithm

Abstract

This research paper presents an innovative approach to cluster the dataset and apply the different machine learning algorithms. For clustering, K-Means is used. Before applying K-Means, estimation of value of K is done by using the Elbow method followed by Silhouette method. For the entire experimental work, the Kaggle dataset has been taken into account. K-Means clustering used over here offers five different clusters. Here, each cluster shows identification of distinct student. Subsequently, four prominent machine learning algorithms—K-Nearest Neighbor (KNN), Neural Network (NN), Random Forest (RF), and Support Vector Machine (SVM) are applied and the performance metrics are measured. The comparative analysis of the machine learning algorithms reveals varying levels of accuracy, precision, recall, and F1-score across different clusters. The outcomes highlight the algorithm that exhibits superior performance in this specific context. Here the highest accuracy i.e. 92.00% is achieved with Random Forest algorithm.