An Empirical Comparison among Supervised Learning Algorithms with Model Explainability

Supervised machine learning algorithms refer to the implicit discovery of significant patterns in data with high accuracy. However, model accuracy does not always fulfill the researcher’s demands. Currently, researchers tend to concentrate their attention on extracting qualitative insights from outputs rather than model accuracy. In this regard, three widely used machine learning algorithms named Decision Tree, Support Vector Machine, and Naive Bayes together with seven benchmark datasets of diverse domains are chosen for this comparative study. The study demonstrates the reasons behind each result to assess each algorithm’s effectiveness. Therefore, the three classifiers are thoroughly examined in terms of several performance indicators, computational complexity, and AUC-ROC curve interpretation to understand the comparison fully. The final result of this study serves as a valuable guide for researchers and practitioners seeking to select the most appropriate machine learning algorithm for their specific data and application requirements.