Indian Journal of Science and Technology
Year: 2023, Volume: 16, Issue: 3, Pages: 204-213
Aditya Nellutla1*, N Srinivasan2
1Research Scholar, Sathyabama University, Chennai, Tamil Nadu, India
2Professor, Department of Computer Science and Engineering, Rajalakshmi Engineering College, Chennai, Tamil Nadu, India
Email: [email protected]
Received Date:06 October 2022, Accepted Date:01 December 2022, Published Date:23 January 2023
Objectives: Utility-list based algorithms have gained a lot of traction due to their efficiency and the ease with which they may be modified. While there have been some enhancements, the problem of inefficiency persists. This research presents a solution to this issue by enhancing the utility-list building process, a crucial function that has received little attention in previous studies. Also, the research aims at reducing memory complexity and better performance than existing approaches. Methods: To expedite building, a new set of bitwise operations termed Bit combine construction (BCC) is proposed. In addition, BCC is supported by a unique data format called EBP (Efficiency Bit Partition). An innovative EBP-Miner algorithm is developed with this framework in mind, and it uses many techniques to narrow the search field. Findings: On widely used baseline methods, experimental findings reveal that EBP-Miner outperforms numerous state-of-the-art techniques, including FEACP as well as CLH-Miner approaches. The experiments were conducted with utilization value ranging from 20% to 100% of the nodes. The proposed system achieves an average of 390s runtime and utilization value of 90.25% which are outperformed the existing methods. Also, the approach has proven 20% lesser memory complexity than of the existing algorithms. Novelty: In the field of data mining, high utility itemset mining (HUIM) is an important challenge. The idea is to discover groups of data in a database that are particularly significant or profitable in order to unearth information that can aid in making decisions. The novelty of this study is on developing a better method for building algorithms for HUIM that make use of a bitwise data structure, and on suggesting a more time- and effort-effective strategy for building utility-lists.
Keywords: High Utility Itemsets; Data Mining; Optimization Model; Bitwise Operations; Pattern Mining
© 2023 Nellutla & Srinivasan. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Published By Indian Society for Education and Environment (iSee)
Subscribe now for latest articles and news.