High-Utility Itemset Mining (HUIM) is an essential task in data mining, focused on identifying itemsets that deliver high utility, such as profit or significance, from extensive datasets. However, traditional HUIM algorithms often grapple with challenges related to scalability and efficiency, particularly when processing large and complex datasets. In order to overcome these constraints, this study suggests a unique HUIM method that was motivated by nature-based optimization techniques. The proposed approach incorporates a hash-based mechanism integrated with Binary Particle Swarm Optimization (HUIM-BPSO), which is particularly advantageous for large datasets or computationally intensive tasks by reducing redundant computations and enhancing execution time. Extensive experiments will be conducted on UCI benchmark datasets to evaluate the algorithm’s robustness. The suggested method’s performance will be evaluated against current methods in a number of different parameters, including the number of high-utility itemsets identified, efficiency, convergence rate, execution time, memory usage, and accuracy. The proposed HUIM-BPSO algorithms use only 0.06 MB memory to process the dataset and uses vary less time ~ 2 ms to provide HUIM. The findings aim to demonstrate significant improvements in computational efficiency and scalability, contributing to the advancement of HUIM methodologies.