E. Omiecinski and A. Sarasere. Efficient mining of association rules in large dynamic databases. In Proc. BNCOD'98, 1998. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Efficient Algorithms for Incremental Update of Frequent.. - Zhang, Kao, Cheung, Yip (2002) (Correct)
....It needs to generate a number of intermediate databases during the process. If main memory is large enough, PrefixSpan is very efficient; otherwise it will require a high cost. Finally, a number of studies have been done on the problem of maintaining discovered association rules including [2, 3, 4, 5, 7, 8, 12, 14]. 4 Algorithms In this section we describe the two incremental update algorithms GSP and MFS . The idea is that, given a sequence s, we use the support count of s in D (if available) to deduce whether s could have enough support in D . In the deduction process, the portion of the database that ....
E. Omiecinski and A. Sarasere. Efficient mining of association rules in large dynamic databases. In Proc. BNCOD'98, 1998.
An Efficient Algorithm To Update Large Itemsets With Early.. - Ayan, Tansel, Arkun (1999) (5 citations) (Correct)
....on the set of whole transactions, i.e. old transactions plus new transactions. However, this process is not efficient since it ignores the previously discovered rules, and repeats all the work done previously. Therefore, algorithms for efficiently updating the association rules were proposed in [7, 8, 12, 14, 16]. These algorithms take the set of association rules in the old database into account, and use this knowledge 1) to remove itemsets that do not exist in the updated database, and 2) to add new rules which were not in the set of old transactions but implied in the updated database. Particularly, ....
....old transactions is large, these algorithms discover the new set of association rules much faster than by re running an algorithm over the whole database. In this paper, we propose an algorithm called UWEP (Update With Early Pruning) that follows the approaches of FUP 2 [8] and Partition Update [12] algorithms. It works iteratively on the new set of transactions, like the previous 2 algorithms. The advantages of UWEP are that it scans the existing database at most once and new database exactly once, and it generates and counts the minimum number of candidates in order to determine the new ....
[Article contains additional citation context not shown here]
Edward Omiecinski and Ashok Savasere. Efficient mining of association rules in large dynamic databases. In Proceedings of BNCOD'98, pages 49--63, 1998.
Updating Large Itemsets with Early Pruning - Ayan (1999) (Correct)
....whole transactions, i.e. old transactions plus new transactions. However, this process is not efficient since it ignores the previously discovered rules, and repeats all the work done previously. Therefore, algorithms for efficiently updating the association rules were proposed in [CHNW96, CLK97, OS98, SS98a, TBAR97] These algorithms take the set of association rules in the old database into account, and use this knowledge 1) to remove itemsets that do not exist in the updated database, and 2) to add new rules which were not in the set of old transactions but implied in the updated database. ....
....is large, these algorithms discover the new set of association rules much faster than by re running an algorithm over the whole database. In this thesis, we propose an algorithm called UWEP (Update With Early Pruning) that follows the approaches of FUP 2 [CLK97] and Partition Update [OS98] algorithms. It works iteratively on the new set of transactions, like the previous algorithms. The advantages of UWEP are that it scans the 31 CHAPTER 3. UPDATING LARGE ITEMSETS 32 existing database at most once and new database exactly once, and it generates and counts the minimum number of ....
[Article contains additional citation context not shown here]
Edward Omiecinski and Ashoka Savasere. Efficient mining of association rules in large dynamic databases. In Proceedings of 16 th British National Conference on Databases (BNCOD'98), pages 49-- 63, Cardiff, Wales, UK, July 1998. BIBLIOGRAPHY 72
An Efficient Algorithm to Update Large Itemsets with Early.. - Ayan, Tansel, Arkun (1999) (5 citations) (Correct)
....solution is to re run an algorithm, say Apriori [2] on the updated database. However, this process is not efficient since it ignores the previously discovered rules, and repeats all the work done previously. Therefore, algorithms for efficiently updating the association rules were proposed in [4, 5, 7, 9, 11]. These algorithms take the set of association rules in the old database into account, and use this knowledge 1) to remove itemsets that no longer exist in updated database, and 2) to add new itemsets which were not in the set of old transactions but now exist in the updated database. In this ....
....that no longer exist in updated database, and 2) to add new itemsets which were not in the set of old transactions but now exist in the updated database. In this paper, we propose an algorithm called UWEP (Update With Early Pruning) that follows the approaches of FUP 2 [5] and Partition Update [7] algorithms, and provides an improvement over them. The advantages of UWEP are that it scans the existing database at most once and new database exactly once, and it generates and counts the minimum number of candidate itemsets in order to determine the new set of large itemsets. Moreover, it ....
[Article contains additional citation context not shown here]
E. Omiecinski and A. Savasere. Efficient mining of association rules in large dynamic databases. In Proc. BNCOD'98, pages 49--63, 1998.
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