Title
Fast Discovery of Minimal Sets of Attributes Functionally Determining a Decision Attribute
Abstract
In our paper, we offer an efficient Funalgorithm for discovering minimal sets of conditional attributes functionally determining a given dependent attribute, and in particular, for discovering Rough Sets certain, generalized decision, and membership distribution reducts. Funcan operate either on partitions or alternatively on stripped partitions that do not store singleton groups. It is capable of using functional dependencies occurring among conditional attributes for pruning candidate dependencies. The experimental results show that all variants of Funhave similar performance. They also prove that Funis much faster than the Rosetta toolkit's algorithms computing all reducts and faster than TANE, which is one of the most efficient algorithms computing all minimal functional dependencies.
Year
DOI
Venue
2007
10.1007/978-3-540-73451-2_34
RSEISP
Keywords
Field
DocType
functional dependency,minimal set,attributes functionally,minimal functional dependency,minimal sets,funhave similar performance,rosetta toolkit,conditional attribute,efficient funalgorithm,dependent attribute,efficient algorithm,membership distribution reducts,fast discovery,decision attribute,rough set
Data mining,Functional dependency,Rough set,Singleton,Mathematics
Conference
Volume
ISSN
Citations 
4585
0302-9743
15
PageRank 
References 
Authors
0.83
13
2
Name
Order
Citations
PageRank
Marzena Kryszkiewicz11662118.72
Piotr Lasek2644.15