Data clustering using hybrid water cycle algorithm and a local pattern search method

•Hybrid water cycle with evaporation rate and Hookes-Jeeves algorithms.•The application of proposed hybrid method for data clustering problems.•The proposed method outperforms other clustering methods reported in literature.•Performance is based on solution quality and/or number of function evaluations. Cluster analysis is a valuable data analysis and data mining technique. Nature-inspired population-based metaheuristics are promising search methods for solving optimization problems including data clustering. In this paper, a recently proposed algorithm called the water cycle algorithm, based on the evaporation rate is used in conjunction with a local search method namely Hookes and Jeeves method to perform data clustering. Statistical analyses were carried out which show that the hybrid optimization method, in general, performs superior to the methods reported in the literature in terms of solution quality as well as computational performance. The proposed hybrid algorithm is tested on some selected standard datasets obtained from the UCI machine-learning repository. The objective function is based on the Euclidean distance as well as the DB index. The experimental results were compared with the data clustering results reported in published literature. The simulation results confirm the superiority of the proposed hybrid method as an efficient and reliable algorithm to solve clustering problems.