Show simple item record

dc.contributor.advisorHamerly, Gregory James, 1977-
dc.contributor.authorFeng, Yu.
dc.contributor.otherBaylor University. Dept. of Computer Science.en
dc.date.accessioned2007-03-19T14:52:48Z
dc.date.available2007-03-19T14:52:48Z
dc.date.copyright2006-12
dc.date.issued2007-03-19T14:52:48Z
dc.identifier.urihttp://hdl.handle.net/2104/5021
dc.descriptionIncludes bibliographical references (p. 50-52).en
dc.description.abstractWe present a novel algorithm called PG-means in this thesis. This algorithm is able to determine the number of clusters in a classical Gaussian mixture model automatically. PG-means uses efficient statistical hypothesis tests on one-dimensional projections of the data and model to determine if the examples are well represented by the model. In so doing, we apply a statistical test to the entire model at once, not just on a per-cluster basis. We show that this method works well in difficult cases such as overlapping clusters, eccentric clusters and high dimensional clusters. PG-means also works well on non-Gaussian clusters and many true clusters. Further, the new approach provides a much more stable estimate of the number of clusters than current methods.en
dc.description.statementofresponsibilityby Yu Feng.en
dc.format.extentvii, 52 p. : ill.en
dc.format.extent193840 bytes
dc.format.extent1477879 bytes
dc.format.mimetypeapplication/pdf
dc.format.mimetypeapplication/pdf
dc.language.isoen_USen
dc.rightsBaylor University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. Contact librarywebmaster@baylor.edu for inquiries about permission.en
dc.subjectAlgorithms.en
dc.subjectComputer network architecture.en
dc.titlePG-means: learning the number of clusters in data.en
dc.typeThesisen
dc.description.degreeM.S.en
dc.rights.accessrightsWorldwide accessen
dc.contributor.departmentComputer Science.en


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record