A Robust Speaker Clustering Algorithm

In this paper, we present a novel speaker segmentation and clustering algorithm. The algorithm automatically performs both speaker segmentation and clustering without any prior knowledge of the identities or the number of speakers. Advantages of this algorithm over other approaches are: no need for training/development data, no threshold adjustment requirements, and robustness to different data conditions. This paper also reports the performance of the algorithm on different datasets released by NIST with different initial conditions and parameter settings. The consistently low speaker diarization error rate clearly indicates the robustness of the algorithm.

Jitendra Ajmera

2003

Fast high-dimensional Bayesian classification and clustering

Vahid Partovi Nia

We introduce a fast approach to classification and clustering applicable to high-dimensional continuous data, based on Bayesian mixture models for which explicit computations are available. This permits us to treat classification and clustering in a single framework, and allows calculation of unobserved class probability. The new classifier is robust to adding noise variables as a drawback of the built-in spike-and-slab structure of the proposed Bayesian model. The usefulness of classification using our method is shown on metabololomic example, and on the Iris data with and without noise variables. Agglomerative hierarchical clustering is used to construct a dendrogram based on the posterior probabilities of particular partitions, to provide a dendrogram with a probabilistic interpretation. An extension to variable selection is proposed which summarises the importance of variables for classification or clustering and has probabilistic interpretation. Having a simple model provides estimation of the model parameters using maximum likelihood and therefore yields a fully automatic algorithm. The new clustering method is applied to metabolomic, microarray, and image data and is studied using simulated data motivated by real datasets. The computational difficulties of the new approach are discussed, solutions for algorithm acceleration are proposed, and the written computer code is briefly analysed. Simulations shows that the quality of the estimated model parameters depends on the parametric distribution assumed for effects, but after fixing the model parameters to reasonable values, the distribution of the effects influences clustering very little. Simulations confirms that the clustering algorithm and the proposed variable selection method is reliable when the model assumptions are wrong. The new approach is compared with the popular Bayesian clustering alternative, MCLUST, fitted on the principal components using two loss functions in which our proposed approach is found to be more efficient in almost every situation.

EPFL2009

Cluster Validity Measure and Merging System for Hierarchical Clustering considering Outliers

Maurice Borgeaud, Frank Grégoire Jean de Morsier, Volker Gass, Jean-Philippe Thiran, Devis Tuia

Clustering algorithms have evolved to handle more and more complex structures. However, measures allowing to qualify the quality of such partitions are rare and only specic to certain algorithms. In this work, we propose a new cluster validity measure (CVM) handling solutions with arbitrary shapes and various levels of outlier rejection based on notions of cluster cores and outliers. Moreover, we propose an adequate cluster merging system (CMS) to group cluster cores sharing some of their outliers. These outliers may be a mixture of these nearby cores. The extension of the Support Vector Clustering and Gaussian Process Clustering to obtain true hierarchical solutions are presented and applied using the proposed CVM and CMS in synthetic and real experiments showing the benefit for hyperparameter selection.

Elsevier2015

Fast high-dimensional Bayesian classification and clustering

Vahid Partovi Nia

EPFL2009