On K-means algorithm with the use of mahalanobis distances

Igor Melnykov, Volodymyr Melnykov

    Research output: Contribution to journalArticle

    40 Citations (Scopus)

    Abstract

    The K-means algorithm is commonly used with the Euclidean metric. While the use of Mahalanobis distances seems to be a straightforward extension of the algorithm, the initial estimation of covariance matrices can be complicated. We propose a novel approach for initializing covariance matrices.

    Original languageEnglish
    Pages (from-to)88-95
    Number of pages8
    JournalStatistics and Probability Letters
    Volume84
    Issue number1
    DOIs
    Publication statusPublished - Jan 2014

    Fingerprint

    Mahalanobis Distance
    K-means Algorithm
    Covariance matrix
    Euclidean
    Metric
    K-means
    Mahalanobis distance

    Keywords

    • Initialization
    • K-means algorithm
    • Mahalanobis distance

    ASJC Scopus subject areas

    • Statistics, Probability and Uncertainty
    • Statistics and Probability

    Cite this

    On K-means algorithm with the use of mahalanobis distances. / Melnykov, Igor; Melnykov, Volodymyr.

    In: Statistics and Probability Letters, Vol. 84, No. 1, 01.2014, p. 88-95.

    Research output: Contribution to journalArticle

    Melnykov, Igor ; Melnykov, Volodymyr. / On K-means algorithm with the use of mahalanobis distances. In: Statistics and Probability Letters. 2014 ; Vol. 84, No. 1. pp. 88-95.
    @article{9c14fbe8020b4bdea8aeacbbea39bcd5,
    title = "On K-means algorithm with the use of mahalanobis distances",
    abstract = "The K-means algorithm is commonly used with the Euclidean metric. While the use of Mahalanobis distances seems to be a straightforward extension of the algorithm, the initial estimation of covariance matrices can be complicated. We propose a novel approach for initializing covariance matrices.",
    keywords = "Initialization, K-means algorithm, Mahalanobis distance",
    author = "Igor Melnykov and Volodymyr Melnykov",
    year = "2014",
    month = "1",
    doi = "10.1016/j.spl.2013.09.026",
    language = "English",
    volume = "84",
    pages = "88--95",
    journal = "Statistics and Probability Letters",
    issn = "0167-7152",
    publisher = "Elsevier",
    number = "1",

    }

    TY - JOUR

    T1 - On K-means algorithm with the use of mahalanobis distances

    AU - Melnykov, Igor

    AU - Melnykov, Volodymyr

    PY - 2014/1

    Y1 - 2014/1

    N2 - The K-means algorithm is commonly used with the Euclidean metric. While the use of Mahalanobis distances seems to be a straightforward extension of the algorithm, the initial estimation of covariance matrices can be complicated. We propose a novel approach for initializing covariance matrices.

    AB - The K-means algorithm is commonly used with the Euclidean metric. While the use of Mahalanobis distances seems to be a straightforward extension of the algorithm, the initial estimation of covariance matrices can be complicated. We propose a novel approach for initializing covariance matrices.

    KW - Initialization

    KW - K-means algorithm

    KW - Mahalanobis distance

    UR - http://www.scopus.com/inward/record.url?scp=84885983328&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84885983328&partnerID=8YFLogxK

    U2 - 10.1016/j.spl.2013.09.026

    DO - 10.1016/j.spl.2013.09.026

    M3 - Article

    AN - SCOPUS:84885983328

    VL - 84

    SP - 88

    EP - 95

    JO - Statistics and Probability Letters

    JF - Statistics and Probability Letters

    SN - 0167-7152

    IS - 1

    ER -