TAILIEUCHUNG - MBIS: An efficient method for mining frequent weighted utility itemsets from quantitative databases
With this structure, the calculation of the intersection of tidsets between two itemsets becomes more convenient. Based on this structure, the authors define the MBiS-Tree structure and propose an algorithm for mining FWUIs from quantitative databases. Experimental results for a number of databases show that the proposed method outperforms existing methods. | Journal of Computer Science and Cybernetics, , (2015), 17–30 DOI: MBIS: AN EFFICIENT METHOD FOR MINING FREQUENT WEIGHTED UTILITY ITEMSETS FROM QUANTITATIVE DATABASES NGUYEN DUY HAM1 , VO DINH BAY2 , NGUYEN THI HONG MINH3 , TZUNG-PEI HONG4 1 Department of Math & Informatics, University of People’s Security, Ho Chi Minh City, Vietnam duyhaman@ 2 Faculty of Information Technology, Ho Chi Minh City University of Technology, Vietnam bayvodinh@ 3 School of Graduate Studies, Vietnam National University, Hanoi, Vietnam minhnth@ 4 Department of Computer Science and Information Engineering, National University of Kaohsiung, Kaohsiung, Taiwan, ROC tphong@ Abstract. In recent years, methods for mining quantitative databases have been proposed. However, the processing time is fairly much, which affects the productivity of intelligent systems that use quantitative databases. This study proposes the multibit segment (MBiS) structure to store and process tidsets to increase the effeciency of mining frequent weighted utility itemsets (FWUIs) from quantitative databases. With this structure, the calculation of the intersection of tidsets between two itemsets becomes more convenient. Based on this structure, the authors define the MBiS-Tree structure and propose an algorithm for mining FWUIs from quantitative databases. Experimental results for a number of databases show that the proposed method outperforms existing methods. Keywords. Dynamic bit vector frequent itemset, frequent weighted utility itemset, multibit segment, tidset 1. INTRODUCTION Mining frequent itemsets (FIs) to find relationships among items plays an important role in data mining, especially for associaiton rules [1] and classification based on association rules [2]. Many algorithms have been proposed to deal with this issue, such as Apriori [1], FP-Growth [3], Charm [4], Eclat [5], and dEclat [6]. These approaches use either a horizontal or .
đang nạp các trang xem trước