Grant: Discovery Grants Program - Individual
University of Guelph
$12,000.00
Topics on model-based clustering
EF 2008-09
Résumé
Clustering means to group, to gather together or to classify. Clustering techniques are used across a wide range of scientific and social disciplines, such as food science, evolution, disease diagnosis and prevention, and in the analysis of bioinformatics data. For example, chemical data can be used to classify a certain food as authentic or not. In recent years, a particular type of clustering; model-based clustering, has become increasingly popular. The work that will be carried out under this proposal is at the cutting-edge of model-based clustering. Existing methods of analyzing data like the food authenticity example given above will be greatly expanded and techniques for the clustering of data that depend on time will be developed. Data that depend on time also arise in a variety of areas. For example, such data could arise from drug trials, where effects are monitored over a period of time.
Work already completed on the construction of a twelve-member family of Gaussian mixture models, with parsimonious factor analysis-like covariance structure, will be replicated for mixtures of t-distributions. This will lead to new family of model-based clustering techniques that will offer the analyst greater modeling diversity. Other work has been carried out using Gaussian mixture models to model longitudinal data will be developed from the existing four models to a larger family of mixture models. This family of models uses a modified Cholesky-decomposed covariance structure and also has the potential for extension to mixtures of t-distributions.
Software will be implemented for each technique so that when this work is complete, a suite of effective and efficient model-based clustering techniques will be available to the analyst. Therefore, these techniques will be accessible to researchers from a broad range of disciplines both here in Canada and abroad.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
356040-2008
Name nom
McNicholas, Paul
Department département
Mathematics and Statistics
Organization id
24
Institution établissement
University of Guelph
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2008
Competition year année de concours
2008
Award amount
12000
Program id
RGPIN
Program nane en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
14
Committee name en
Statistical Sciences
Committee name fr
Statistique
Area of application code
1200
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Advancement of knowledge
Area of application fr
Avancement des connaissances
Research subject code
3011
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Multivariate analysis
Research subject fr
Analyse multivariée
Partie
1
Nb partie
5
Application title
Topics on model-based clustering
Keyword
NA
Application summary
Clustering means to group, to gather together or to classify. Clustering techniques are used across a wide range of scientific and social disciplines, such as food science, evolution, disease diagnosis and prevention, and in the analysis of bioinformatics data. For example, chemical data can be used to classify a certain food as authentic or not. In recent years, a particular type of clustering; model-based clustering, has become increasingly popular. The work that will be carried out under this proposal is at the cutting-edge of model-based clustering. Existing methods of analyzing data like the food authenticity example given above will be greatly expanded and techniques for the clustering of data that depend on time will be developed. Data that depend on time also arise in a variety of areas. For example, such data could arise from drug trials, where effects are monitored over a period of time.
Work already completed on the construction of a twelve-member family of Gaussian mixture models, with parsimonious factor analysis-like covariance structure, will be replicated for mixtures of t-distributions. This will lead to new family of model-based clustering techniques that will offer the analyst greater modeling diversity. Other work has been carried out using Gaussian mixture models to model longitudinal data will be developed from the existing four models to a larger family of mixture models. This family of models uses a modified Cholesky-decomposed covariance structure and also has the potential for extension to mixtures of t-distributions.
Software will be implemented for each technique so that when this work is complete, a suite of effective and efficient model-based clustering techniques will be available to the analyst. Therefore, these techniques will be accessible to researchers from a broad range of disciplines both here in Canada and abroad.