Title :
Compressing the Set of Frequent Sequential Patterns
Author_Institution :
Coll. of Comput. Sci. & Technol., Hubei Univ. of Econ., Wuhan
Abstract :
Compressing the set of frequent sequential patterns is a method in order to address the problem of explosive number of output sequential patterns. In order to get high-quality compression, it first clusters frequent sequential patterns, and then select and output only a representative sequential pattern for each cluster such that the number of these representative sequential patterns is minimized. A greedy algorithm and an efficient candidate-based algorithm are proposed. The set of representative sequential patterns is a kind of subset of frequent sequential patterns. Experimental results show that it can achieve very good compression effect.
Keywords :
data compression; data mining; greedy algorithms; pattern clustering; candidate-based algorithm; frequent sequential pattern compression; frequent sequential pattern mining; greedy algorithm; Clustering algorithms; Computer science; Databases; Educational institutions; Explosives; Fuzzy systems; Greedy algorithms; Itemsets; Proposals; Data Mining; Representative Sequential Pattern; Sequential Pattern;
Conference_Titel :
Fuzzy Systems and Knowledge Discovery, 2008. FSKD '08. Fifth International Conference on
Conference_Location :
Jinan Shandong
Print_ISBN :
978-0-7695-3305-6
DOI :
10.1109/FSKD.2008.168