DNA, life’s architectural design, is made up of long, repeating sequences of nucleic acids that contain the genetic information of living things. The method of obtaining information from DNA is an important part of genetic research. This field’s primary duties are DNA sequencing, which determines the order of base pairs, and DNA sequence classification, which determines if an unlabeled sequence belongs to a recognized class. This paper delves into several machine learning techniques for classifying DNA sequences. These techniques are evaluated using three public datasets. This study used machine learning methods, including multinomial Naive Bayes, to distinguish infected genes from normal genes using labeling and k-mer encoding. The effectiveness of the models was assessed using Naive Bayes multinomial classifications with k-mer encoding, which showed the best accuracy of the experimental data, with an accuracy of 98%, 99%, and 92%, respectively.