Traditional clustering methods assume that there is no measurement error, or uncertainty, associated with data. Often, however, real world applications require treatment of data that have such errors. In the presence of measurement errors, well-known clustering methods like k-means and hierarchical clustering may not produce satisfactory results. The fundamental question addressed in this paper...

Abstract: Distance or similarity measures are of fundamental importance to pattern classification, clustering, and information retrieval problems. Various distance/similarity measures that are applicable to compare two nominal type histograms are reviewed and categorized in both syntactic and semantic relationships. A correlation coefficient and a hierarchical clustering technique are adopted t...

Preserving the privacy of individuals when data are shared for clustering is a complex problem. The challenge is how to protect the underlying data values subjected to clustering without jeopardizing the similarity between objects under analysis. In this short paper, we revisit a family of geometric data transformation methods (GDTMs) that distort numerical attributes by translations, scalings,...

introduction thunderstorms pose a significant threat to modern societies and their assets. despite their local-scale characteristics, severe thunderstorms and associated extreme events like heavy rainfall, hail, gusts, or tornadoes can cause considerable damage to agriculture, buildings, or infrastructure, and facilities. thunderstorms are highly localized and largely stationary weather systems...

We determine the exact freezing threshold, r , for a family of models of random boolean constraint satisfaction problems, including NAE-SAT and hypergraph 2-colouring, when the constraint size is sufficiently large. If the constraint-density of a random CSP, F , in our family is greater than r then for almost every solution of F , a linear number of variables are frozen, meaning that their colo...

Model-based clustering typically involves the development of a family of mixture models and the imposition of these models upon data. The best member of the family is then chosen using some criterion and the associated parameter estimates lead to predicted group memberships, or clusterings. This paper describes the extension of the mixtures of multivariate t-factor analyzers model to include co...

This paper presents a new Similarity Based Agglomerative Clustering(SBAC) algorithm that works well for data with mixed numeric and nominal features. A similarity measure, proposed by Goodall for biological taxonomy[13], that gives greater weight to uncommon feature-value matches in similarity computations and makes no assumptions of the underlying distributions of the feature-values, is adopte...

Family clustering of varying levels of serum PBB and DDE was evaluated for 62 Michigan families by using the mean and standard deviation for family units, ranked percentile comparison of family members, and correlation by linear regression of family members. The results indicate that levels of serum PBBs cluster within family units and for children within families. Serum DDE clustering within f...

This paper proposes a novel algorithm of hierarchical divisive clustering, which generates a multi-branch tree, not a binary one, as its output. In order to use the algorithm for clustering large document sets, a spherical kmeans clustering algorithm based on a cosine measure is adopted for partitioning recursively the document set from the top to bottom. Also, by selecting automatically the nu...