We address the problem of robust clustering by combining data partitions (forming a clustering ensemble) produced by multiple clusterings. We formulate robust clustering under an information-theoretical framework; mutual information is the underlying concept used in the definition of quantitative measures of agreement or consistency between data partitions. Robustness is assessed by variance of the cluster membership, based on bootstrapping. We propose and analyze a voting mechanism on pairwise associations of patterns for combining data partitions. We show that the proposed technique attempts to optimize the mutual information based criteria, although the optimality is not ensured in all situations. This evidence accumulation method is demonstrated by combining the well-known Kmeans algorithm to produce clustering ensembles. Experimental results show the ability of the technique to identify clusters with arbitrary shapes and sizes.

For analyzing shapes of planar, closed curves, we propose di#erential geometric representations of curves using their direction functions and curvature functions. Shapes are represented as elements of infinite-dimensional spaces and their pairwise di#erences are quantified using the lengths of geodesics connecting them on these spaces. We use a Fourier basis to represent tangents to the shape spaces and then use a gradient-based shooting method to solve for the tangent that connects any two shapes via a geodesic.

the results of multiple clusterings. Initially, n d-dimensional data is decomposed into a large number of compact clusters; the K-means algorithm performs this decomposition, with several clusterings obtained by N random initializations of the K-means. Taking the cooccurrences of pairs of patterns in the same cluster as votes for their association, the data partitions are mapped into a co-association matrix of patterns. This n n matrix represents a new similarity measure between patterns. The final clusters are obtained by applying a MST-based clustering algorithm on this matrix. Results on both synthetic and real data show the ability of the method to identify arbitrary shaped clusters in multidimensional data.

In this paper, we identify two issues involved in developing an automated feature subset selection algorithm for unlabeled data: the need for finding the number of clusters in conjunction with feature selection, and the need for normalizing the bias of feature selection criteria with respect to dimension. We explore the feature selection problem and these issues through FSSEM (Feature Subset Selection using Expectation-Maximization (EM) clustering) and through two different performance criteria for evaluating candidate feature subsets: scatter separability and maximum likelihood. We present proofs on the dimensionality biases of these feature criteria, and present a cross-projection normalization scheme that can be applied to any criterion to ameliorate these biases. Our experiments show the need for feature selection, the need for addressing these two issues, and the effectiveness of our proposed solutions.

Abstract—The analysis of a feature space that exhibits multiscale patterns often requires kernel estimation techniques with locally adaptive bandwidths, such as the variable-bandwidth mean shift. Proper selection of the kernel bandwidth is, however, a critical step for superior space analysis and partitioning. This paper presents a mean shift-based approach for local bandwidth selection in the multimodal, multivariate case. Our method is based on a fundamental property of normal distributions regarding the bias of the normalized density gradient. We demonstrate that, within the large sample approximation, the local covariance is estimated by the matrix that maximizes the magnitude of the normalized mean shift vector. Using this property, we develop a reliable algorithm which takes into account the stability of local bandwidth estimates across scales. The validity of our theoretical results is proven in various space partitioning experiments involving the variable-bandwidth mean shift. Index Terms—Variable-bandwidth mean shift, bandwidth selection, multiscale analysis, Jensen-Shannon divergence, feature space. 1

Clustering is a common unsupervised learning technique used to discover group structure in a set of data. While there exist many algorithms for clustering, the important issue of feature selection, that is, what attributes of the data should be used by the clustering algorithms, is rarely touched upon. Feature selection for clustering is difficult because, unlike in supervised learning, there are no class labels for the data and, thus, no obvious criteria to guide the search. Another important problem in clustering is the determination of the number of clusters, which clearly impacts and is influenced by the feature selection issue. In this paper, we propose the concept of feature saliency and introduce an expectation-maximization (EM) algorithm to estimate it, in the context of mixture-based clustering. Due to the introduction of a minimum message length model selection criterion, the saliency of irrelevant features is driven toward zero, which corresponds to performing feature selection. The criterion and algorithm are then extended to simultaneously estimate the feature saliencies and the number of clusters.

Abstract—In this paper, based on ideas from lossy data coding and compression, we present a simple but effective technique for segmenting multivariate mixed data that are drawn from a mixture of Gaussian distributions, which are allowed to be almost degenerate. The goal is to find the optimal segmentation that minimizes the overall coding length of the segmented data, subject to a given distortion. By analyzing the coding length/rate of mixed data, we formally establish some strong connections of data segmentation to many fundamental concepts in lossy data compression and rate-distortion theory. We show that a deterministic segmentation is approximately the (asymptotically) optimal solution for compressing mixed data. We propose a very simple and effective algorithm that depends on a single parameter, the allowable distortion. At any given distortion, the algorithm automatically determines the corresponding number and dimension of the groups and does not involve any parameter estimation. Simulation results reveal intriguing phase-transition-like behaviors of the number of segments when changing the level of distortion or the amount of outliers. Finally, we demonstrate how this technique can be readily applied to segment real imagery and bioinformatic data. Index Terms—Multivariate mixed data, data segmentation, data clustering, rate distortion, lossy coding, lossy compression, image segmentation, microarray data clustering. 1

Clustering ensembles have emerged as a powerful method for improving both the robustness and the stability of unsupervised classification solutions. However, finding a consensus clustering from multiple partitions is a difficult problem that can be approached from graph-based, combinatorial or statistical perspectives. We offer a probabilistic model of consensus using a finite mixture of multinomial distributions in a space of clusterings. A combined partition is found as a solution to the corresponding maximum likelihood problem using the EM algorithm. The excellent scalability of this algorithm and comprehensible underlying model are particularly important for clustering of large datasets. This study compares the performance of the EM consensus algorithm with other fusion approaches for clustering ensembles. We also analyze clustering ensembles with incomplete information and the effect of missing cluster labels on the quality of overall consensus. Experimental results demonstrate the effectiveness of the proposed method on large real-world datasets.

Using a recently proposed geometric representation of planar shapes, we present algorithmic tools for: (i) hierarchical clustering of imaged objects according to the shapes of their boundaries, (ii) learning of probability models for clustered shapes, and (iii) testing of observed shapes under competing probability models. Clustering at any level of hierarchy is performed using a mimimum dispersion criterion and a Markov search process. Statistical means of clusters provide shapes to be clustered at the next higher level, thus building a hierarchy of shapes.

A fundamental problem in computer vision and pattern recognition is to determine where and, most importantly, why a given technique is applicable. This is not only necessary because it helps us decide which techniques to apply at each given time. Knowing why current algorithms cannot be applied, facilitates the design of new algorithms robust to such problems. In this paper, we report on a theoretical study that demonstrates where and why generalized eigen-based linear equations do not work. In particular, we show that when the smallest angle between the i th eigenvector given by the metric to be maximized and the first i eigenvectors given by the metric to be minimized is close to zero, our results are not guaranteed to be correct. Several properties of such models are also presented. For illustration, we concentrate on the classical applications of classification and feature extraction. We also show how we can use our findings to design more robust algorithms. We conclude with a discussion on the broader impacts of our results. Index terms: feature extraction, generalized eigenvalue decomposition, performance evaluation, classifiers, pattern recognition. 1