Clustering constrained symbolic data

• Published in: Pattern recognition letters Vol. 30; no. 11; pp. 1037 - 1045
• Main Authors: de Carvalho, Francisco de A.T., Csernel, Marc, Lechevallier, Yves
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01-08-2009; Elsevier
• Subjects: Applied sciences; Clustering algorithms; Constraints; Dissimilarity functions; Exact sciences and technology; Information, signal and communications theory; Normal symbolic form; Signal and communications theory; Signal representation. Spectral analysis; Signal, noise; Symbolic Data Analysis; Telecommunications and information theory; Symbolic Data Analysis; Dissimilarity functions; Constraints; Normal symbolic form; Clustering algorithms; Data analysis; Automatic classification; Similarity; Normal form; Algorithm; Signal classification; Polynomial time; Computation time; Combinatorial method; Database
• ISSN: 0167-8655; 1872-7344
• DOI: 10.1016/j.patrec.2009.04.009

Dealing with multi-valued data has become quite common in both the framework of databases as well as data analysis. Such data can be constrained by domain knowledge provided by relations between the variables and these relations are expressed by rules. However, such knowledge can introduce a combinatorial increase in the computation time depending on the number of rules. In this paper, we present a way to cluster such data in polynomial time. The method is based on the following: a decomposition of the data according to the rules, a suitable dissimilarity function and a clustering algorithm based on dissimilarities.