Random projection towards the Baire metric for high dimensional clustering

Fionn Murtagh, Pedro Contreras

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

For high dimensional clustering and proximity finding, also referred to as high dimension and low sample size data, we use random projection with the following principle. With the greater probability of close-to-orthogonal projections, compared to orthogonal projections, we can use rank order sensitivity of projected values. Our Baire metric, divisive hierarchical clustering, is of linear computation time.

Original languageEnglish
Title of host publicationStatistical Learning and Data Sciences
Subtitle of host publicationThird International Symposium, SLDS 2015, Egham, UK, April 20-23, 2015, Proceedings
EditorsAlexander Gammerman, Vladimir Vovk, Harris Papadopoulos
PublisherSpringer Verlag
Pages424-431
Number of pages8
ISBN (Electronic)9783319170916
ISBN (Print)9783319170909
DOIs
Publication statusPublished - 3 Apr 2015
Externally publishedYes
Event3rd International Symposium on Statistical Learning and Data Sciences - University of London, Egham, United Kingdom
Duration: 20 Apr 201523 Apr 2015
Conference number: 3
http://www.clrc.rhul.ac.uk/slds2015/ (Link to Conference Website)

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume9047
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference3rd International Symposium on Statistical Learning and Data Sciences
Abbreviated titleSLDS 2015
CountryUnited Kingdom
CityEgham
Period20/04/1523/04/15
Internet address

    Fingerprint

Cite this

Murtagh, F., & Contreras, P. (2015). Random projection towards the Baire metric for high dimensional clustering. In A. Gammerman, V. Vovk, & H. Papadopoulos (Eds.), Statistical Learning and Data Sciences : Third International Symposium, SLDS 2015, Egham, UK, April 20-23, 2015, Proceedings (pp. 424-431). (Lecture Notes in Computer Science; Vol. 9047). Springer Verlag. https://doi.org/10.1007/978-3-319-17091-6_37