Subspace clustering of high-dimensional data: a predictive approach

DOI10.1007/S10618-013-0317-YzbMATH Open1294.68119arXiv1203.1065OpenAlexW2025391886MaRDI QIDQ740601FDOQ740601

Authors: Brian McWilliams, Giovanni Montana 

Publication date: 3 September 2014

Published in: Data Mining and Knowledge Discovery (Search for Journal in Brave)

Abstract: In several application domains, high-dimensional observations are collected and then analysed in search for naturally occurring data clusters which might provide further insights about the nature of the problem. In this paper we describe a new approach for partitioning such high-dimensional data. Our assumption is that, within each cluster, the data can be approximated well by a linear subspace estimated by means of a principal component analysis (PCA). The proposed algorithm, Predictive Subspace Clustering (PSC) partitions the data into clusters while simultaneously estimating cluster-wise PCA parameters. The algorithm minimises an objective function that depends upon a new measure of influence for PCA models. A penalised version of the algorithm is also described for carrying our simultaneous subspace clustering and variable selection. The convergence of PSC is discussed in detail, and extensive simulation results and comparisons to competing methods are presented. The comparative performance of PSC has been assessed on six real gene expression data sets for which PSC often provides state-of-art results.

Full work available at URL: https://arxiv.org/abs/1203.1065

Recommendations

• High-dimensional data clustering
• The generic subspace clustering model
• Subspace clustering with automatic feature grouping
• Partition clustering of high dimensional low sample size data based on \(p\)-values
• Bayesian variable selection in clustering high-dimensional data with substructure

zbMATH Keywords

model selectionvariable selectionPCAsubspace clusteringmicroarraysPRESS statistics

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Learning and adaptive systems in artificial intelligence (68T05) Pattern recognition, speech recognition (68T10) Protein sequences, DNA sequences (92D20)

Cites Work

• A framework for feature selection in clustering
• Estimating the number of clusters in a data set via the gap statistic
• Pathwise coordinate optimization
• Title not available (Why is that?)
• Sharp Thresholds for High-Dimensional and Noisy Sparsity Recovery Using $\ell _{1}$-Constrained Quadratic Programming (Lasso)
• Adjustment of an Inverse Matrix Corresponding to a Change in One Element of a Given Matrix
• Sparse principal component analysis via regularized low rank matrix approximation
• Title not available (Why is that?)
• Title not available (Why is that?)
• Consensus clustering: A resampling-based method for class discovery and visualization of gene expression microarray data
• A survey on enhanced subspace clustering
• Influential observations, high leverage points, and outliers in linear regression
• \(k\)-plane clustering
• Hybrid linear modeling via local best-fit flats
• Asymptotic convergence analysis of the projection approximation subspace tracking algorithms
• Identifying multiple influential observations in linear regression
• Subspace clustering of high-dimensional data: a predictive approach

Cited In (16)

• Partition clustering of high dimensional low sample size data based on \(p\)-values
• Robust subspace clustering
• Analyzing gene expression data with predictive clustering trees
• Is-ClusterMPP: clustering algorithm through point processes and influence space towards high-dimensional data
• Dense community detection in multi-valued attributed networks
• Projective clustering based on Parzen window technique
• Lossy compression approach to subspace clustering
• Subspace clustering for panel data with interactive effects
• Subspace clustering of high-dimensional data: a predictive approach
• Weighted sparse simplex representation: a unified framework for subspace clustering, constrained clustering, and active learning
• High dimensional data clustering from a dynamical systems point of view
• Hierarchical subspace clustering
• Subspace clustering with automatic feature grouping
• Enhancing principal direction divisive clustering
• Influential features PCA for high dimensional clustering
• The generic subspace clustering model

Uses Software

• sparcl

This page was built for publication: Subspace clustering of high-dimensional data: a predictive approach

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q740601)