Boosting Moth-Flame Optimization Algorithm by Arithmetic Optimization Algorithm for Data Clustering

Data clustering (DC) is one of the common data mining problems in which the given objects are grouped into a specific number of clusters. Each cluster contains similar objects, and different clusters contain dissimilar objects. In this research, a novel hybrid clustering method is proposed to solve several common DC problems. The proposed method combined the Moth-Flame Optimization (MFO) algorithm with Arithmetic Optimization Algorithm (AOA), called MFOAOA. The main aim of the proposed MFOAOA is to incorporate the advantages of the exploration search ability of the MFO algorithm and the exploitation search of the AOA. The proposed MFOAOA is tested using several benchmark DC problems and compared with several well-known methods, including AOA, Particle Swarm Optimizer, Gray Wolf Optimizer, Sine Cosine Algorithm, Aquila Optimizer (AO), Whale Optimization Algorithm, and MFO algorithm. The results showed that the proposed MFOAOA obtained better results in all the tested clustering-based approaches.