DNA Sequence Classification with Improved Performance of Supervised Classifiers using Nature Inspired Algorithms

Deoxyribo Nucleic Acid (DNA) is a unique macromolecule of all living species. It passes on the hereditary data of life. Artificial Intelligence can be used to classify the DNA class based on the sequence, as it can reduce the time taken and can enhance the information managing limits compared to a manual technique that is followed for ages. The main objective of this project is to develop a machine learning model that can improve the classification of DNA sequences. Firefly Algorithm (FA), Grey Wolf Optimization (GWO) Algorithm, Bat Algorithm (BA), and Hybrid Bat Algorithm (HBA) were some of the Nature Inspired Algorithms (NIA) have been employed in this research. These NIAs generally follow “Evolutionary Computation”. Six different supervised classifiers are utilized to classify the DNA sequence. This research is a practical survey aimed at identifying the best classifier with the best NIA among the candidates for reliably classifying DNA sequences. With the use of NIAs, all of the classifier's performance has improved, however, GWO's random forest classifier has the highest accuracy of all of the classifiers tested at 97.1 percent.