. "Dolov\u00E1n\u00ED siln\u00FDch vzor\u016F z l\u00E9ka\u0159sk\u00FDch sekven\u010Dn\u00EDch dat"@cs . . "4" . "Mining the Strongest Patterns in Medical Sequential Data"@en . "Dolov\u00E1n\u00ED siln\u00FDch vzor\u016F z l\u00E9ka\u0159sk\u00FDch sekven\u010Dn\u00EDch dat" . "\u017Delezn\u00FD, Filip" . "Mining the Strongest Patterns in Medical Sequential Data"@en . "data mining; sequential data"@en . . . "L\u00E9ka\u0159 a technika" . "CZ - \u010Cesk\u00E1 republika" . "Karel, Filip" . . . "Holas, T." . . . "3"^^ . "Sekven\u010Dn\u00ED data jsou d\u016Fle\u017Eit\u00FDm zdrojem l\u00E9ka\u0159sk\u00FDch znalost\u00ED. Tato specifick\u00E1 data mohou vznikat \u0159adou r\u016Fzn\u00FDch zp\u016Fsob\u016F. V tomto \u010Dl\u00E1nku na p\u0159\u00EDkladu konkr\u00E9tn\u00ED studie prezentujeme obecn\u00E9 postupy pro jejich dolov\u00E1n\u00ED. Jde o preventivn\u00ED dlouhodobou studii atheroskler\u00F3zy - data jsou v\u00FDsledkem dv\u011B dek\u00E1dy trvaj\u00EDc\u00EDho sledov\u00E1n\u00ED v\u00FDvoje rizikov\u00FDch faktor\u016F a p\u0159idru\u017Een\u00FDch jev\u016F. Hlavn\u00EDm c\u00EDlem je identifikovat \u010Dast\u00E9 sekven\u010Dn\u00ED vzory, tj. opakuj\u00EDc\u00ED se \u010Dasov\u00E9 jevy, a studovat jejich mo\u017Enou souvislost s objeven\u00EDm jedn\u00E9 ze sledovan\u00FDch kardiovaskul\u00E1rn\u00EDch nemoc\u00ED. Z \u0161ir\u0161\u00ED \u0161k\u00E1ly dostupn\u00FDch metod se soust\u0159ed\u00EDme na induktivn\u00ED logick\u00E9 programov\u00E1n\u00ED, kter\u00E9 potenci\u00E1ln\u00ED vzory vyjad\u0159uje ve form\u011B rys\u016F v predik\u00E1tov\u00E9 logice prvn\u00EDho \u0159\u00E1du. Rysy jsou nejprve automaticky extrahov\u00E1ny a n\u00E1sledn\u011B sdru\u017Eov\u00E1ny do pravidel, kter\u00E1 p\u0159edstavuj\u00ED v\u00FDstupn\u00ED formu z\u00EDskan\u00E9 znalosti. Navr\u017Een\u00FD postup je porovn\u00E1n s tradi\u010Dn\u011Bj\u0161\u00EDmi metodami publikovan\u00FDmi d\u0159\u00EDve. Jde o metodu posuvn\u00FDch oken a epizodn\u00ED pravidla."@cs . "Kl\u00E9ma, Ji\u0159\u00ED" . "Dolov\u00E1n\u00ED siln\u00FDch vzor\u016F z l\u00E9ka\u0159sk\u00FDch sekven\u010Dn\u00EDch dat"@cs . "4"^^ . . "0301-5491" . "364165" . "Sekven\u010Dn\u00ED data jsou d\u016Fle\u017Eit\u00FDm zdrojem l\u00E9ka\u0159sk\u00FDch znalost\u00ED. Tato specifick\u00E1 data mohou vznikat \u0159adou r\u016Fzn\u00FDch zp\u016Fsob\u016F. V tomto \u010Dl\u00E1nku na p\u0159\u00EDkladu konkr\u00E9tn\u00ED studie prezentujeme obecn\u00E9 postupy pro jejich dolov\u00E1n\u00ED. Jde o preventivn\u00ED dlouhodobou studii atheroskler\u00F3zy - data jsou v\u00FDsledkem dv\u011B dek\u00E1dy trvaj\u00EDc\u00EDho sledov\u00E1n\u00ED v\u00FDvoje rizikov\u00FDch faktor\u016F a p\u0159idru\u017Een\u00FDch jev\u016F. Hlavn\u00EDm c\u00EDlem je identifikovat \u010Dast\u00E9 sekven\u010Dn\u00ED vzory, tj. opakuj\u00EDc\u00ED se \u010Dasov\u00E9 jevy, a studovat jejich mo\u017Enou souvislost s objeven\u00EDm jedn\u00E9 ze sledovan\u00FDch kardiovaskul\u00E1rn\u00EDch nemoc\u00ED. Z \u0161ir\u0161\u00ED \u0161k\u00E1ly dostupn\u00FDch metod se soust\u0159ed\u00EDme na induktivn\u00ED logick\u00E9 programov\u00E1n\u00ED, kter\u00E9 potenci\u00E1ln\u00ED vzory vyjad\u0159uje ve form\u011B rys\u016F v predik\u00E1tov\u00E9 logice prvn\u00EDho \u0159\u00E1du. Rysy jsou nejprve automaticky extrahov\u00E1ny a n\u00E1sledn\u011B sdru\u017Eov\u00E1ny do pravidel, kter\u00E1 p\u0159edstavuj\u00ED v\u00FDstupn\u00ED formu z\u00EDskan\u00E9 znalosti. Navr\u017Een\u00FD postup je porovn\u00E1n s tradi\u010Dn\u011Bj\u0161\u00EDmi metodami publikovan\u00FDmi d\u0159\u00EDve. Jde o metodu posuvn\u00FDch oken a epizodn\u00ED pravidla." . . "[9A2669AA8DAD]" . . . . "Dolov\u00E1n\u00ED siln\u00FDch vzor\u016F z l\u00E9ka\u0159sk\u00FDch sekven\u010Dn\u00EDch dat" . "38" . "RIV/68407700:21230/08:03155677!RIV09-AV0-21230___" . "P(1ET101210513)" . . . "8"^^ . "Sequential data represent an important source of automatically mined and potentially new medical knowledge. They can originate in various ways. Within the presented domain they come from a longitudinal preventive study of atherosclerosis - the data consists of series of long-term observations recording the development of risk factors and associated conditions. The intention is to identify frequent sequential patterns having any relation to an onset of any of the observed cardiovascular diseases. This paper focuses on application of inductive logic programming. The prospective patterns are based on first-order features automatically extracted from the sequential data. The features are further grouped in order to reach final complex patterns expressed as rules. The presented approach is also compared with the approaches published earlier (windowing, episode rules)."@en . "21230" . . "RIV/68407700:21230/08:03155677" .