Peningkatan Algoritma Naïve Bayes Menggunakan Algoritma Genetika Pada Klasifikasi Bakteri

Abstract

In previous studies, only using the nave Bayes algorithm and resulted in an accuracy value of 80.93% and currently the accuracy value will be increased by using a genetic algorithm. Learning patterns in genetic algorithms can inform new entrants or new classifications with a faster time. Difficulties in the configuration that exist in nave Bayes can be helped by this genetic algorithm in addition to being able to provide adequate modeling to describe the system. Bacteria consists of three classifications of bacteria: based on how they obtain food, based on gram staining, and based on their shape. In this bacterial data, which consists of numerical parameters containing the sequence name, Mcg, gvh, Lip, chg, aac, alm1, alm2 and a class distributron of 336 attributes, in the class distribution there are 8 protein classes classified, namely cytoplasm (cp), membrane inside without signal sequence (im), perisplasm (pp), inside membrane with uncleavable signal sequence (IMU), outer membrane (om), outer membrane lipoprotein (OML), inside membrane lipoprotein (IML), inner membrane with cleavable signal sequence ( STI). The results of this study resulted in an accuracy value of 81.19%. Thus, it is proven that the performance of the nave Bayes algorithm can be improved by using a genetic algorithm for bacterial classification