TY - GEN
T1 - Discovery of generalized association rules with multiple minimum supports
AU - Lui, Chung Leung
AU - Chung, Fu Lai Korris
PY - 2000/1/1
Y1 - 2000/1/1
N2 - Mining association rules at multiple concept levels leads to the discovery of more concrete knowledge. Nevertheless, setting an appropriate minsup for cross-level itemsets is still a non-trivial task. Besides, the mining process is computationally expensive and produces many redundant rules. In this work, we propose an efficient algorithm for mining generalized association rules with multiple minsup. The method scans appropriately k+1 times of the number of original transactions and once of the extended transactions where k is the largest itemset size. Encouraging simulation results were reported.
AB - Mining association rules at multiple concept levels leads to the discovery of more concrete knowledge. Nevertheless, setting an appropriate minsup for cross-level itemsets is still a non-trivial task. Besides, the mining process is computationally expensive and produces many redundant rules. In this work, we propose an efficient algorithm for mining generalized association rules with multiple minsup. The method scans appropriately k+1 times of the number of original transactions and once of the extended transactions where k is the largest itemset size. Encouraging simulation results were reported.
UR - http://www.scopus.com/inward/record.url?scp=84875251797&partnerID=8YFLogxK
M3 - Conference article published in proceeding or book
SN - 9783540410669
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 510
EP - 515
BT - Principles of Data Mining and Knowledge Discovery - 4th European Conference, PKDD 2000, Proceedings
PB - Springer Verlag
T2 - 4th European Conference on Principles and Practice of Knowledge Discovery in Databases, PKDD 2000
Y2 - 13 September 2000 through 16 September 2000
ER -