Next statistics seminar at the University of Edinburgh:
Note DIFFERENT TIME
Friday 13th February *4 p.m.*
JCMB Room 3317
NEMA DEAN, University of Glasgow
Mixture Model Component Cluster Trees
Abstract:
One of the most commonly used parametric clustering methods -
model-based clustering - assumes that continuous data (possibly after a
transformation) comes from a mixture of Gaussian components. The common
implicit assumption is that once the best such mixture has been chosen
to fit the data, each mixture component is a cluster estimating an
underlying (subpopulation) group. Clearly there will be issues with such
an assumption if the underlying groups do not have Gaussian
distributions. While the mixture will still fit the data well, it is
likely that if the true underlying groups are non-symmetric, skewed,
heavy-tailed, curvilinear or if there are outliers then the number of
components in the model will overestimate the number of groups. We look
at using hierarchical clustering methods based on a distance defined by
the estimated mixture to create a dendrogram with components as leaves -
a component cluster tree. This can be used to identify submixtures of
combinations of components that will better estimate the underlying groups.
Seminars website: http://www.maths.ed.ac.uk/seminars/list/series/7
--
Dr Natalia Bochkina
Lecturer in Statistics
School of Mathematics
King’s Buildings
University of Edinburgh
Mayfield Road
Edinburgh, EH9 3JZ
Tel: 0131 650 8597
Email: [log in to unmask]
Webpage: http://www.maths.ed.ac.uk/~nbochkin/
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
|