A high utility itemsets mining algorithm based on co-evolution

Metaheuristic high utility itemsets mining algorithms often face challenges such as poor initial population quality, low time efficiency, and itemsets loss due to premature convergence. To address these issues, this study proposes a high utility itemsets mining algorithm based on co-evolution. A population initialization strategy based on logarithmic decay and probability distribution is proposed to enhance population diversity and improve the quality of initial solutions. Additionally, to improve search efficiency and computational performance, a co-evolutionary update strategy is designed, where particle swarm optimization enhances the Lévy flight mechanism in cuckoo search. Furthermore, an adaptive simplified mutation strategy is introduced to increase population diversity and convergence speed, thereby reducing the risk of itemset loss. Experimental results show that the proposed algorithm outperforms state-of-the-art methods in terms of the number of high utility itemsets mined, runtime, recall, precision, and convergence.