HDT-HS: A hybrid decision tree/harmony search algorithm for biological datasets

This paper introduces the Hybrid Decision Tree with Harmony Search (HDT-HS) optimization algorithm to improve the rate of accuracy for the decision tree algorithm so as to apply it to DNA data sets. The hybridization includes operating the decision tree method after the Improvisation step of the harmony search algorithm in order to navigate for several solutions at the same time. This is to improve the accuracy of the final results for the decision tree. The results show that the hybrid algorithm achieved better accuracy of about 96.73% compared to classifier algorithms such as Nave (94.8%), MBBC (95.99%); optimization algorithms such as bagging (94.5%) and boosting (94.7%); hybrid decision tree with genetic algorithm (70.7%) and another version from the decision tree such as C4.5 (94.3%) and PCL (94.4%).