N. Ayan, A. Tansel, and E. Arkun. An Ecient Algorithm to Update Large Itemsets with Early Pruning. In Proc. of the 5th ACM-SIGKDD Intl. Conf. on Knowledge Discovery and Data Mining, August 1999.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....algorithm on the updated database or with the new parameters. However, this process is not ecient since it is memoryless (i.e. it ignores the already discovered knowledge) essentially duplicating part of the work that have already been done. To address this problem, several researchers [3, 4, 6, 9, 10, 16, 17] have proposed incremental association mining algorithms. These algorithms re use the previously mined information and try to combine this information with the fresh data to eciently re compute the new set of association rules. In this paper we present a novel technique that advances the ....

....in incremental association rule mining. Our algorithm, called zigzag 1 , like other approaches, uses previously discovered knowledge to reduce the cost of updating the frequent itemsets. However, it introduces some signi cant improvements over previous incremental mining algorithms [3, 4, 6]. We highlight these improvements next. Our approach maintains only the maximal frequent itemsets to incrementally construct the lattice of frequent itemsets. The maximal frequent itemsets are updated by a novel backtracking search, which is guided by the results of the previous mining iteration, ....

[Article contains additional citation context not shown here]

N. Ayan, A. Tansel, and E. Arkun. An Ecient Algorithm to Update Large Itemsets with Early Pruning. In Proc. of the 5th ACM-SIGKDD Intl. Conf. on Knowledge Discovery and Data Mining, August 1999.

....and removing the restriction that it learn exactly the tree that a batch system would. A comparison between BOAT and CVFDT is an area for future work. There has been a great deal of work on incrementally maintaining association rules. Cheung, Han, Ng, and Wong [7] and Fazil, Tansel, and Arkun [2] propose algorithms for maintaining sets of association rules when new transactions are added to the database. Sarda and Srinivas [22] have also done some work in the area. DEMON s contribution [11] is particularly relevant, as it addresses association rule maintenance speci cally in the ....

N. F. Ayan, A. U. Tansel, and M. E. Arkun. An ecient algorithm to update large itemsets with early pruning. In Proceedings of the Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 287-291, San Diego, CA, 1999. ACM Press.

No context found.

N. Ayan, A. Tansel, and E. Arkun. An Ecient Algorithm to Update Large Itemsets with Early Pruning. In Proc. of the 5th ACM-SIGKDD Intl. Conf. on Knowledge Discovery and Data Mining, August 1999.

No context found.

N. Ayan, A. Tansel, and E. Arkun. An Ecient Algorithm to Update Large Itemsets with Early Pruning. In Proc. of the 5th ACM-SIGKDD Intl. Conf. on Knowledge Discovery and Data Mining, August 1999.

No context found.

N. F. Ayan, A. U. Tansel, and E. Arkun. An ecient algorithm to update large itemsets with early pruning. In Proc. of the 5th Int'l Conf. on Knowledge Discovery and Data Mining (KDD '99), San Diego, California, USA, August 1999.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC