In this paper, we propose a model-based clustering method (TVClust) that robustly incorporates noisy side information as soft-constraints and aims to seek a consensus between side information and the observed data. Our method is based on a nonparametric Bayesian hierarchical model that combines the probabilistic model for the data instance and the one for the side-information. An efficient Gibb...

Usually, incremental algorithms for data streams clustering not only suffer from sensitive initialization parameters, but also incorrectly represent large classes by many cluster representatives, which leads to decrease the computational efficiency over time. We propose in this paper an incremental clustering algorithm based on ”growing neural gas” (GNG), which addresses this issue by using a p...

My research work is centered around clustering, the most basic form of unsupervised learning. It extends to several closely related areas including: probabilistic mixture models, manifold leaning and dimension reduction, low rank approximation, maximum margin clustering and semi-supervised learning. An important theme of my work is to scale up the currently existing, but computationally demandi...

Data clustering is a challenging task in data mining technique. Various clustering algorithms are developed to cluster or categorize the datasets. Many algorithms are used to cluster the categorical data. Some algorithms cannot be directly applied for clustering of categorical data. Several attempts have been made to solve the problem of clustering categorical data via cluster ensembles. But th...

This paper presents a diffusion based probabilistic interpretation of spectral clustering and dimensionality reduction algorithms that use the eigenvectors of the normalized graph Laplacian. Given the pairwise adjacency matrix of all points, we define a diffusion distance between any two data points and show that the low dimensional representation of the data by the first few eigenvectors of th...

Classical model-based partitional clustering algorithms, such as k-means or mixture of Gaussians, provide only loose and indirect control over the size of the resulting clusters. In this work, we present a family of probabilistic clustering models that can be steered towards clusters of desired size by providing a prior distribution over the possible sizes, allowing the analyst to fine-tune exp...

The aim is to facilitate the application of user defined constraints to the genetic clustering algorithm. This is achieved by presenting a general penalty function. The penalty function is defined as a normal distribution. The function is augmented to an extensible environment to assemble genetic clustering algorithms, called DAGC. The main idea behind the design of DAGC is to provide the resea...