ESO-HUIM: Enhanced Swarm Optimization for High-Utility Item-Set Mining in Critical Product Identification from Transaction Logs

In recent times, High-Utility Item-Set Mining (HUIM) has garnered significant attention for its ability to identify successful products by simultaneously considering both quantity and profit, diverging from traditional methods such as frequent item-set mining or association rule mining. Addressing the challenges posed by vast item sets and sizable databases, utility-driven mining has emerged as a robust data mining approach, seeking meaningful insights from diverse datasets.This research introduces a novel swarm-based algorithm for mining High-Utility Item-Sets (HUIs). The algorithm is characterized by unique features that demonstrate its effectiveness in extracting high-utility data from large datasets. Beyond the algorithmic contribution, this thesis includes the application of the proposed method to mine high-utility data from real-world datasets, adding empirical evidence to its practical relevance.The presented research significantly advances the field of High-Utility Mining by developing highly efficient algorithms and showcasing their practical applicability. The emphasis on the usefulness of items underscores the importance of deriving valuable insights from data. The title "ESO-HUIM: Enhanced Swarm Optimization for High-Utility Item-Set Mining in Critical Product Identification from Transaction Logs" encapsulates the essence of this research, highlighting the algorithm's enhanced swarm optimization approach for critical product identification through HUIM in transaction logs.