Mining frequent closed cubes in 3D datasets
VLDB '06 Proceedings of the 32nd international conference on Very large data bases
Mining association rules in very large clustered domains
Information Systems
CSV: visualizing and mining cohesive subgraphs
Proceedings of the 2008 ACM SIGMOD international conference on Management of data
Mining long high utility itemsets in transaction databases
WSEAS Transactions on Information Science and Applications
Top-down mining of frequent closed patterns from very high dimensional data
Information Sciences: an International Journal
Finding closed frequent item sets by intersecting transactions
Proceedings of the 14th International Conference on Extending Database Technology
Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery
Hi-index | 0.00 |
The problem of mining frequent closed patterns has receivedconsiderable attention recently as it promises to have much lessredundancy compared to discovering all frequent patterns. Existingalgorithms can presently be separated into two groups,feature (column) enumeration and row enumeration. Featureenumeration algorithms like CHARM and CLOSET+ areefficient for datasets with small number of features and largenumber of rows since the number of feature combinations to beenumerated is small. Row enumeration algorithms like CARPENTERon the other hand are more suitable for datasets (eg.bioinformatics data) with large number of features and smallnumber of rows. Both groups of algorithms, however, will encounterproblem for datasets that have large number of rowsand features.In this paper, we describe a new algorithm called COBBLERwhich can efficiently mine such datasets . COBBLER isdesigned to dynamically switch between feature enumerationand row enumeration depending on the data characteristic inthe process of mining. As such, each portion of the datasetcan be processed using the most suitable method, making themining more efficient. Several experiments on real-life andsynthetic datasets show that COBBLER is an order of magnitudebetter than previous closed pattern mining algorithmslike CHARM, CLOSET+ and CARPENTER.