Mining Top-K High Utility Itemsets Using Bio-Inspired Algorithms with a Diversity within Population Framework

High-utility itemset mining (HUIM), as a necessary data mining task, has paid the attention of many researchers. It includes numerous applications in various arears. Recently, a method, which improved the memory usage and runtime for HUIs mining, was proposed, is called TKO-BPSO. It helps to automatically increase the border thresholds and might considerably reduce the combinational problem for pruning the search space effectively. However, the idea only works to maintain the current optimal values in the next populations, leading to the variety within populations is limited. To handle this problem, we propose a new bio-inspired algorithm-based HUIM framework to explore HUIs, namely TKO-HUIMF-PSO (Top-K high utility itemset mining in One phase based on a HUIM Framework of Particle Swarm Optimization). The main idea of TKO-HUIMF-PSO adapts the standard roadmap of bio-inspired algorithms by applying roulette wheel selection to all the discovered HUIs to determine the target values of the next population. Consequently, it improves the diversity within populations. Significant experiments conducted on publicly available several real and synthetic datasets delineate that the proposed algorithm is efficient and effective in terms of runtime and memory usage.