Number of the records: 1  

Measures of Ruleset Quality Capable to Represent Uncertain Validity

  1. 1.
    0089741 - ÚI 2008 RIV DE eng C - Conference Paper (international conference)
    Holeňa, Martin
    Measures of Ruleset Quality Capable to Represent Uncertain Validity.
    [Míry kvality souborů pravidel schopné reprezentovat nejistou platnost.]
    Symbolic and Quantitative Approaches to Reasoning with Uncertainty. Berlin: Springer, 2007 - (Mellouli, K.), s. 430-442. Lecture Notes in Artificial Intelligence, 4724. ISBN 978-3-540-75255-4.
    [ECSQARU 2007. European Conference /9./. Hammamet (TN), 01.11.2007-02.11. 2007]
    R&D Projects: GA ČR GA201/05/0325
    Institutional research plan: CEZ:AV0Z10300504
    Keywords : rules extraction from data * quality measures * Boolean rules * fuzzy rules * observational logic * classification * ROC curves
    Subject RIV: IN - Informatics, Computer Science

    The paper deals with quality measures of rulesets. Three particular approaches to the generalization such measures from classification to general rulesets are discussed. One of them, capable to represent uncertain validity, is elaborated in some detail.

    Článek se zabývá mírami kvality souborů pravidel. Jsou diskutovány tři konkrétní přístupy ke zobecnění takových měr z klasifikačních na obecné soubory pravidel. Jeden z nich, schopný reprezentovat nejistou platnost, je rozpracován detailněji.
    Permanent Link: http://hdl.handle.net/11104/0150849

     
     
Number of the records: 1  

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