This work centers on a novel data mining technique we term supervised clustering. Unlike traditional clustering, supervised clustering assumes that the examples are classified and has the goal of identifying class-uniform clusters that have high probability densities. Three representative–based algorithms for supervised clustering are introduced: two greedy algorithms SRIDHCR and SPAM, and an e...

A popular graph clustering method is to consider the embedding of an input graph into R induced by the first k eigenvectors of its Laplacian, and to partition the graph via geometric manipulations on the resulting metric space. Despite the practical success of this methodology, there is limited understanding of several heuristics that follow this framework. We provide theoretical justification ...

In this paper, we propose a new clustering model for speaker diarization. A major problem with using greedy agglomerative hierarchical clustering for speaker diarization is that they do not guarantee an optimal solution. We propose a new clustering model, by redefining clustering as a problem of Integer Linear Programming (ILP). Thus an ILP solver can be used which searches the solution of spea...

Discovering protein sequence motif information is one of the most crucial tasks in bioinformatics research. In this paper, we try to obtain protein recurring patterns which are universally conserved across protein family boundaries. In order to achieve the goal, our dataset is extremely large. Therefore, an efficient technique is required. In this article, short recurring segments of proteins a...

in a strapdown magnetic compass, heading angle is estimated using the earth's magnetic field measured by three-axis magnetometers (tam). however, due to several inevitable errors in the magnetic system, such as sensitivity errors, non-orthogonal and misalignment errors, hard iron and soft iron errors, measurement noises and local magnetic fields, there are large error between the magnetometers'...

This work centers on a novel data mining technique we term supervised clustering. Unlike traditional clustering, supervised clustering assumes that the examples are classified and has the goal of identifying class-uniform clusters that have high probability densities. Three representative–based algorithms for supervised clustering are introduced: two greedy algorithms SRIDHCR and SPAM, and an e...

We introduce a new Bayesian model for hierarchical clustering based on a prior over trees called Kingman’s coalescent. We develop novel greedy and sequential Monte Carlo inferences which operate in a bottom-up agglomerative fashion. We show experimentally the superiority of our algorithms over the state-of-the-art, and demonstrate our approach in document clustering and phylolinguistics.

A method for structural clustering is proposed involving data on subset-to-entity linkages that can be calculated with structural data such as graphs or sequences or images. The method is based on the layered structure of the problem of maximization of a set function de ned as the minimum value of linkages between a set and its elements and referred to as the tightness function. When the linkag...