作者: Santosh Kumar , G. Sahoo
DOI: 10.1007/978-981-10-3874-7_77
关键词: Computer science 、 Genetic algorithm 、 Feature selection 、 Diabetes diagnosis 、 Heart disease 、 Machine learning 、 Artificial intelligence
摘要: This paper proposes a hybrid technique that applies artificial bee colony (ABC) algorithm for the feature selection and combined k-nearest neighbor (k-NN) with genetic (GA) used effective classification. The aim of this was to select finest features including elimination insignificant datasets severely affect classification accuracy. proposed approach in heart disease diabetes diagnosis, which has higher impact rate on reducing quality life throughout world, is developed. disease, diabetes, hepatitis are taken from UCI repository evaluated by technique. accuracy achieved 10-fold cross-validation. Experimental results show our compared other existing systems.