This research presents a comparative analysis on performance of different types of Naïve Bayes classifiers, mainly, Gaussian, Multinomial, and Bernoulli classifier, using different types of datasets—diabetes, heart disease, breast cancer, and textual datasets. The classifiers performance was evaluated using standard estimation metrics, including accuracy, recall, F1-score, and precision. The uniqueness of the proposed research is the point that it does an in-depth analysis on various types of NB classifier, and thus the strengths and weaknesses of each classifier, depending on the dataset's characteristics, are revealed. The results demonstrate how the choice of Naïve Bayes classifier significantly affects the performance. This study aims toward providing insights on the suitability of each classifier for different types of data, aiding in more informed decision-making using machine learning classifiers.