Active Constrained Clustering by Examining Spectral Eigenvectors
نویسندگان
چکیده
This work focuses on the active selection of pairwise constraints for spectral clustering. We develop and analyze a technique for Active Constrained Clustering by Examining Spectral eigenvectorS (ACCESS) derived from a similarity matrix. The ACCESS method uses an analysis based on the theoretical properties of spectral decomposition to identify data items that are likely to be located on the boundaries of clusters, and for which providing constraints can resolve ambiguity in the cluster descriptions. Empirical results on three synthetic and five real data sets show that ACCESS significantly outperforms constrained spectral clustering using randomly selected constraints.
منابع مشابه
Constrained Spectral Clustering under a Local Proximity Structure Assumption
This work focuses on incorporating pairwise constraints into a spectral clustering algorithm. A new constrained spectral clustering method is proposed, as well as an active constraint acquisition technique and a heuristic for parameter selection. We demonstrate that our constrained spectral clustering method, CSC, works well when the data exhibits what we term local proximity structure. Empiric...
متن کاملActive Query Selection and Spectral Eigenvectors Semi-Supervised Clustering
Semi-supervised clustering aims to improve clustering performance by considering user supervision in the form of pairwise constraints. In this paper, we study the active learning problem of selecting pairwise must-link and cannot-link constraints for semisupervised clustering. We consider active learning in an iterative manner where in each iteration queries are selected based on the current cl...
متن کاملSpectral clustering, kernel density estimation and its application on ischemia detection
We considered new ways of clustering, classification and segmentation using spectral graph theory. We devised a novel method of estimating the number of clusters. From this, we could contrive a graph partitioning procedure which used physical intuitions of mass-spring network. And we used the technique of estimating the number of clusters to approximate probability density. This allowed us to s...
متن کاملRobust and Efficient Computation of Eigenvectors in a Generalized Spectral Method for Constrained Clustering
FAST-GE is a generalized spectral method for constrained clustering [Cucuringu et al., AISTATS 2016]. It incorporates the mustlink and cannot-link constraints into two Laplacian matrices and then minimizes a Rayleigh quotient via solving a generalized eigenproblem, and is considered to be simple and scalable. However, there are two unsolved issues. Theoretically, since both Laplacian matrices a...
متن کاملSpectral clustering with eigenvector selection based on entropy ranking
Ng–Jordan–Weiss (NJW) method is one of the most widely used spectral clustering algorithms. For a K clustering problem, this method partitions data using the largest K eigenvectors of the normalized affinity matrix derived from the dataset. It has been demonstrated that the spectral relaxation solution of K-way grouping is located on the subspace of the largest K eigenvectors. However, we find ...
متن کامل