A comparison of unsupervised dimension reduction algorithms for classification

Jaegul Choo, Hyunsoo Kim, Haesun Park, Hongyuan Zha

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

    1 Citation (Scopus)

    Abstract

    Distance preserving dimension reduction (DPDR) using the singular value decomposition has recently been introduced. In this paper, for disease diagnosis using gene or protein expression data, we present empirical comparison results between DPDR and other various dimension reduction (DR) methods (i.e. PCA, MDS, Isomap, and LLE) when using support vector machines with radial basis function kernel. Our results show that DPDR outperforms, as a whole, other DR methods in terms of classification accuracy, but at the same time, it gives significant efficiency compared with other methods since it has no parameter to be optimized. Based on these empirical results, we reach a promising conclusion that DPDR is one of the best DR methods at hand for modeling an efficient and distortion-free classifier for gene or protein expression data.

    Original languageEnglish
    Title of host publicationProceedings - 2007 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2007
    Pages71-77
    Number of pages7
    DOIs
    Publication statusPublished - 2007
    Event2007 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2007 - Fremont, CA, United States
    Duration: 2007 Nov 22007 Nov 4

    Publication series

    NameProceedings - 2007 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2007

    Conference

    Conference2007 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2007
    Country/TerritoryUnited States
    CityFremont, CA
    Period07/11/207/11/4

    ASJC Scopus subject areas

    • Biotechnology
    • General Computer Science
    • Biomedical Engineering

    Fingerprint

    Dive into the research topics of 'A comparison of unsupervised dimension reduction algorithms for classification'. Together they form a unique fingerprint.

    Cite this