Number of the records: 1  

Features of Neighbors Spaces

  1. 1.
    SYSNO ASEP0103283
    Document TypeC - Proceedings Paper (int. conf.)
    R&D Document TypeConference Paper
    TitleFeatures of Neighbors Spaces
    TitleVlastnosti prostoru sousedů
    Author(s) Jiřina, Marcel (UIVT-O) SAI, RID
    Jiřina jr., M. (CZ)
    Source TitleSOFSEM 2004: Theory and Practice of Computer Science. - Berlin : SpringerVerlag, 2004 / Van Emde Boa P. ; Pokorný J. ; Bieliková M. ; Štuller J. - ISBN 3-540-20779-1
    Pagess. 241-248
    Number of pages8 s.
    ActionSOFSEM 2004. Conference on Current Trends in Theory and Practice of Computer Science /30./
    Event date24.01.2004-30.01.2004
    VEvent locationMěřín
    CountryCZ - Czech Republic
    Event typeWRD
    Languageeng - English
    CountryDE - Germany
    Keywordsdistribution mapping function ; power application ; nearest neighbor ; boundary effect
    Subject RIVBA - General Mathematics
    R&D ProjectsLN00B096 GA MŠMT - Ministry of Education, Youth and Sports (MEYS)
    UT WOS000189212700020
    EID SCOPUS35048814461
    DOI https://doi.org/10.1007/978-3-540-24618-3_20
    AnnotationDistances of the nearest neighbor or several nearest neighbors are essential in probability density estimate by the method of k nearest neighbors or in problems of searching in large databases. A typical task of the probability density estimate using several nearest neighbors is the Bayes s classifier. The task of searching in large databases is looking for other nearest neighbor queries. In this paper it is shown that for a uniform distribution of points in an n-dimensional Euclidean space the distribution of the distance of the i-th nearest neighbor to the n-power has Erlang distribution. The power approximation of the newly introduced probability distribution mapping function of distances of nearest neighbors in the form of suitable power of the distance is presented. A way to state distribution mapping exponent q for a probability density estimation including boundary effect in high dimensions is shown.
    WorkplaceInstitute of Computer Science
    ContactTereza Šírová, sirova@cs.cas.cz, Tel.: 266 053 800
    Year of Publishing2005
Number of the records: 1  

  This site uses cookies to make them easier to browse. Learn more about how we use cookies.