Dissimilarity representations allow for building good classifiers

EM Pekalska, RPW Duin

    Research output: Contribution to journalArticleScientificpeer-review

    170 Citations (Scopus)


    In this paper, a classification task on dissimilarity representations is considered. A traditional way to discriminate between objects represented by dissimilarities is the nearest neighbor method. It suffers, however, from a number of limitations, i.e., high computational complexity, a potential loss of accuracy when a small set of prototypes is used and sensitivity to noise. To overcome these shortcomings, we propose to use a normal density-based classifier constructed on the same representation. We show that such a classifier, based on a weighted combination of dissimilarities, can significantly improve the nearest neighbor rule with respect to the recognition accuracy and computational effort. Keywords: Similarity representations; Normal density-based classifiers
    Original languageUndefined/Unknown
    Pages (from-to)943-956
    Number of pages14
    JournalPattern Recognition Letters
    Issue number8
    Publication statusPublished - 2002


    • academic journal papers
    • ZX CWTS JFIS < 1.00

    Cite this