Dimension reduction can produce visualizations of hierarchical structures, like those produced by cluster analysis. So far, reliability of such visualizations has only been assessed with rudimentary means. Here, a method for assessing reliability of such visualizations is developed. It measures how accurately the location of a data point in high-dimensional hierarchy tree can be inferred from a...

Mixtures of Gaussian factors are powerful tools for modeling an unobserved heterogeneous population, offering at the same time dimension reduction and model-based clustering. The high prevalence of spurious solutions and the disturbing effects of outlying observations in maximum likelihood estimation may cause biased or misleading inferences. Restrictions for the component covariances are consi...

In this study, we present a wavelet-based method for removing noise from images and a Bayesian shrinkage factor was derived to estimate noise-free wavelet coefficients. This method took into account dependencies between wavelet coefficients. The interscale dependencies were measured from the local singularity and a conditional probability model was proposed. The intrascale dependencies were mea...

We present a new implementation of the cluster expansion formalism. The new code, UNiversal CLuster Expansion (UNCLE), consolidates recent advances in the methodology and leverages one new development in the formalism itself. As a core goal, the package reduces the need for user intervention, automating the method to reduce human error and judgment. The package extends standard cluster expansio...

To deal with the problem of insufficient labeled data, usually side information – given in the form of pairwise equivalence constraints between points – is used to discover groups within data. However, existing methods using side information typically fail in cases with high-dimensional spaces. In this paper, we address the problem of learning from side information for high-dimensional data. To...

The semantic representation of individual word senses and concepts is of fundamental importance to several applications in Natural Language Processing. To date, concept modeling techniques have in the main based their representation either on lexicographic resources, such as WordNet, or on encyclopedic resources, such as Wikipedia. We propose a vector representation technique that combines the ...

Denoising autoencoder (DAE) is effective in restoring clean speech from noisy observations. In addition, it is easy to be stacked to a deep denoising autoencoder (DDAE) architecture to further improve the performance. In most studies, it is supposed that the DAE or DDAE can learn any complex transform functions to approximate the transform relation between noisy and clean speech. However, for l...

Recorded votes in legislative bodies are an important source of data for political scientists. Voting records can be used to describe parliamentary processes, identify ideological divides between members and reveal the strength of party cohesion. We explore the problem of working with vote data using popular dimensionality reduction techniques and cluster validation methods, as an alternative t...

AIMS The community-based AERLI intervention provided training and education to people who inject drugs (PWID) about HIV and HCV transmission risk reduction, with a focus on drug injecting practices, other injection-related complications, and access to HIV and HCV testing and care. We hypothesized that in such a population where HCV prevalence is very high and where few know their HCV serostatus...

Exploring multivariate spatial data attracts much attention in the visualization community. The main challenge lies in that automatic analysis techniques is insufficient in discovering complicated patterns with the perspective of human beings, while visualization techniques are incapable of accurately identifying the features of interest. This paper addresses this contradiction by enhancing aut...