Mining of Closed High-Utility Itemsets in Dynamic and Incremental Databases

Abstract

The strategy in this study relates to Closed High Utility Itemset mining in dynamic and incremental databases, with focus on efficiency enhancement and adaptability. The investigation was conducted with respect to tree hierarchies, vertical header lists, transaction-weighted utilization, subtree utility, and updating mechanisms for optimizing the outcomes of CHUI mining. A quantitative method was used for the validation of five hypotheses, proving substantial improvements in runtime, scalability, and accuracy of data. Results will highlight an urgent need for dynamic frameworks in AI-driven decision-making and data analysis. Future work should consider expanded diverse database environments and alternative strategies in response to emerging trends in utility mining.