Mining top-k closed itemsets using best-first search

Songram, Panida; Boonjing, Veera

doi:10.1109/CIT.2008.4594653

Mining top-k closed itemsets using best-first search

Author(s)

Songram, Panida

Boonjing, Veera

Date Issued

September 22, 2008

Type

Conference Paper

DOI

10.1109/CIT.2008.4594653

Abstract

In this paper, we proposed an efficient algorithm, TOPK_CLOSED, for mining top-k closed itemsets. This algorithm mines top-k closed itemsets using best-first search. The closed itemsets with the highest supports are firstly found from this algorithm. Consequently, the method quickly finds top-k closed itemsets, which leads to an efficient pruning unnecessary itemsets and stop mining rapidly. In addition, this algorithm can generate closed itemsets without keeping candidates in main memory. © 2008 IEEE.

Citation

Proceedings 2008 IEEE 8th International Conference on Computer and Information Technology CIT 2008, 77-82, 2008