Probabilistic Finite Mixture Clustering of Genetic Expression Microarray Data

Exploratory cluster analysis of large data sets often implements k-means or hierarchical methods. These routines typically exhibit limitations which reduces the reliability of the results. Both assign observations to the one component for which its Euclidean distance from the center is smallest. Furthermore, these methods force a common variance in every dimension and do not permit covariances between dimensions. A final limitation is that these methods prohibit statistical inference.

Grantor dc:publisher

Wake Forest University

Year dc:date.issued

2013

Author dc:creator

• Higgins, Ixavier

Language dc:language.iso

en

Handle dc:identifier.uri

http://hdl.handle.net/10339/38565

OAI identifier oai:identifier

oai:wakespace.lib.wfu.edu:10339/38565