Automatic subspace clustering of high dimensional data for data mining applications

Abstract

Data mining applications place special requirements on clustering algorithms including: the ability to find clusters embedded in subspaces of high dimensional data, scalability, end-user comprehensibility of the results, non-presumption of any canonical data distribution, and insensitivity to the order of input records. We present CLIQUE, a clustering algorithm that satisfies each of these requirements. CLIQUE identifies dense clusters in subspaces of maximum dimensionality. It generates cluster descriptions in the form of DNF expressions that are minimized for ease of comprehension. It produces identical results irrespective of the order in which input records are presented and does not presume any specific mathematical form for data distribution. Through experiments, we show that CLIQUE efficiently finds accurate cluster in large high dimensional datasets.

Keywords

Computer scienceCluster analysisLinear subspaceCliqueData miningScalabilityClustering high-dimensional dataTheoretical computer scienceCluster (spacecraft)Artificial intelligenceDatabaseMathematicsProgramming language

Affiliated Institutions

IBM Research - Almaden US

Related Publications

Swarm: robust and fast clustering method for amplicon-based studies

Frédéric Mahé , Torbjørn Rognes , Christopher Quince +2 more

Popular de novo amplicon clustering methods suffer from two fundamental flaws: arbitrary global clustering thresholds, and input-order dependency induced by centroid selection. ...

2014 PeerJ 907 citations

Model-Based Clustering, Discriminant Analysis, and Density Estimation

Chris Fraley , Adrian E. Raftery

Cluster analysis is the automated search for groups of related observations in a dataset. Most clustering done in practice is based largely on heuristic but intuitively reasonab...

2002 Journal of the American Statistical A... 4130 citations

Search and clustering orders of magnitude faster than BLAST

R. C. Edgar

Abstract Motivation: Biological sequence data is accumulating rapidly, motivating the development of improved high-throughput methods for sequence classification. Results: UBLAS...

2010 Bioinformatics 20899 citations

Stability-Based Validation of Clustering Solutions

Tilman Lange , Volker Röth , Mikio L. Braun +1 more

Data clustering describes a set of frequently employed techniques in exploratory data analysis to extract “natural” group structure in data. Such groupings need to be validated ...

2004 Neural Computation 508 citations

Kernel k-means

Inderjit S. Dhillon , Yuqiang Guan , Brian Kulis

Kernel k-means and spectral clustering have both been used to identify clusters that are non-linearly separable in input space. Despite significant research, these methods have ...

2004 1184 citations

Publication Info

Year: 1998
Type: article
Volume: 27
Issue: 2
Pages: 94-105
Citations: 656
Access: Closed

External Links

View on DOI.org

Social Impact

Altmetric

Automatic subspace clustering of high dimensional data for data mining applications

PlumX Metrics

Social media, news, blog, policy document mentions

Citation Metrics

656

OpenAlex

Cite This

APA Style

                            
                                
                                    Rakesh Agrawal, 
                                
                                    Johannes Gehrke, 
                                
                                    Dimitrios Gunopulos
                                
                                et al.
                            
                            (1998). 
                            Automatic subspace clustering of high dimensional data for data mining applications. 
                            ACM SIGMOD Record
                            , 27
                            (2)
                            , 94-105.
                            https://doi.org/10.1145/276305.276314
                        

Identifiers

DOI: 10.1145/276305.276314