Abstract
Data clustering is one of the challenges of machine learning problems that group a set of data objects into a subset of a predefined number of groups. This paper proposes a new, improved version of the reptile search algorithm (RSA) called quantum mutation reptile search algorithm (QMRSA). The proposed method uses the quantum mutation-based search strategy to enhance the performance of the RSA to solve various optimization problems. The method tackles the main shortcomings raised in the original version of the RSA, like premature convergence and non-equilibrium between the search processes. Experiments are conducted on several benchmark functions and data clustering problems. The results are analyzed and compared with several stateof-the-art methods, including aquila optimizer, grey wolf optimizer, sine cosine algorithm, whale optimization algorithm, dragonfly algorithm, and arithmetic optimization algorithm. The results show the QMRSA???s superiority in dealing with the mathematical benchmark functions and real-world problems like data clustering.