Adaptive optimization with reinforcement learning for high utility itemset extraction

Title

Adaptive optimization with reinforcement learning for high utility itemset extraction

Creator

Logeswaran, K.; Suresh, P.; Savitha, S.; Anandamurugan, S.

Description

Extraction of High Utility Itemsets (HUI) plays a vital role in data mining that comprises several techniques developed to address it efficiently. However, when dealing with large itemsets and diverse items in a dataset, the problem's search space becomes notably complex and expansive. This makes the task of identifying HUIs more computationally expensive and time-consuming. In this paper, a novel Optimized Coverage list unit utilities-based High Utility Itemset (OCHUI) extraction approach is introduced for High Utility Itemset extraction. The extraction of high utility patterns and the extraction of qualified high utility itemsets are the two main processes in the suggested method. In the first step, high utility patterns are identified by mining metrics such as Redefined transaction-weighted utility, positive and negative Unit profit, Purchase quantity, and Coverage (RUPC) from the dataset. In the second step, qualified high utility itemsets are obtained optimally using an adaptive optimization algorithm called Cuckoo search Assisted Ant colony Optimization with Reinforcement Learning (CAAO-RL) is proposed. The Reinforcement Learning (RL) uses the On and Off policy method to intelligently leverage the tuning parameter of optimization. The RUPC model obtained the pattern score of 13600, runtime rate of 10.256 s and memory usage of 198 MB, respectively. 2025 Elsevier B.V.

Source

Knowledge-Based Systems;Volume;331;Issue;;Article No.;114733;

Date

01-01-2025

Publisher

Elsevier B.V.

Subject

High utility itemset mining; Optimization algorithm; Optimized coverage list; Reinforcement learning; RUPC

Coverage

Logeswaran K., Department of AI, ML and Data Science, School of Engineering and Technology, CHRIST University, Kengeri, Karnataka, Bangalore, 560074, India; Suresh P., Department of Database Systems, School of Computer Science and Engineering, Vellore Institute of Technology, Tamil Nadu, Vellore, 632 014, India; Savitha S., Department of Computer Science and Engineering, K.S.R College of Engineering, Tamil Nadu, Tiruchengode, 637215, India; Anandamurugan S., Department of Information Technology, Kongu Engineering College, Perundurai, Erode, Tamilnadu, 638060, India

Rights

Restricted Access; Hardcopy may be available in the library

Relation

ISSN: 9507051; CODEN: KNSYE

Format

online

Language

English

Type

Article

Identifier

https://doi.org/10.1016/j.knosys.2025.114733

https://www.scopus.com/pages/publications/105020986688?origin=resultslist