Kernel logistic regression using truncated Newton method

Maher Maalouf, Theodore B. Trafalis, Indra Adrianto

    Research output: Contribution to journalArticlepeer-review

    26 Scopus citations

    Abstract

    Kernel logistic regression (KLR) is a powerful nonlinear classifier. The combination of KLR and the truncated-regularized iteratively re-weighted least-squares (TR-IRLS) algorithm, has led to a powerful classification method using small-to-medium size data sets. This method (algorithm), is called truncated-regularized kernel logistic regression (TR-KLR). Compared to support vector machines (SVM) and TR-IRLS on twelve benchmark publicly available data sets, the proposed TR-KLR algorithm is as accurate as, and much faster than, SVM and more accurate than TR-IRLS. The TR-KLR algorithm also has the advantage of providing direct prediction probabilities.

    Original languageBritish English
    Pages (from-to)415-428
    Number of pages14
    JournalComputational Management Science
    Volume8
    Issue number4
    DOIs
    StatePublished - Nov 2011

    Keywords

    • Classification
    • Kernel methods
    • Logistic regression
    • Truncated Newton method

    Fingerprint

    Dive into the research topics of 'Kernel logistic regression using truncated Newton method'. Together they form a unique fingerprint.

    Cite this