<p>Clustering data into coherent grouping is a machine learning unsupervised technique that aims to partition the data objects into clusters with maximum inter-cluster distance and minimum intra-cluster distance. Genetic Algorithm is a popular metaheuristic algorithm with several shortcomings, such as local optima trapping, premature convergence, and genetic drift problems. Consequently, to maintain a balance between intensification and diversification and to resolve the above issues, a new memetic algorithm is proposed in this paper. The proposed approach (HGASA) is a hybrid of genetic and simulated annealing algorithms that explores the neighborhood search strategy of the simulated annealing algorithm by incorporating it as a local search technique after the mutation phase of the genetic algorithm. To overwhelm the genetic drift problem, the proposed algorithm's mutation rate is also adjusted using a self-adaptive factor that relies on the metaheuristic search principle. The proposed technique is validated by comparing its performance with a Genetic Algorithm using standard benchmark functions. Afterward, the proposed HGASA is simulated on nine real-life and synthetic clustering data sets. The efficiency of the proposed hybrid approach is proved by comparing it with well-known standard metaheuristics and recently published hybrid algorithms using the sum of intra-cluster distance and convergence speed as objective functions. The effectiveness of the proposed algorithm is also tested as an automatic clustering algorithm. Computational findings reveal that the HGASA clustering technique performs better than other metaheuristic algorithms. Simulation results demonstrate the ability of the proposed HGASA to produce compact and well-separated clusters. Additionally, the Friedman statistical test and two paired t-tests are performed to verify the suggested HGASA method's existence, and findings approve the survival of the HGASA algorithm in the clustering area.</p>