作者: Sonia Mota , Eduardo Ros , Francisco de Toro , Julio Ortega
DOI: 10.1007/3-540-44864-0_105
关键词: Task (project management) 、 Data mining 、 Genetic algorithm 、 Computer science 、 Line (geometry) 、 Paroxysmal atrial fibrillation 、 Set (abstract data type)
摘要: Paroxysmal Atrial Fibrillation (PAF) prediction viability is a line of research currently being investigated. The definition new valid parameters for this task may generate various heterogeneous features. Genetic Algorithms (GAs) automatically find set to maximize the diagnosis capabilities scheme based on K-nearest neighbours algorithm. This an efficient way generating number possible solutions problem PAF prediction. present paper illustrates how GAs, rather than statistical study database can be used select giving best classification rates.