作者: M. Sudha
DOI: 10.1007/S10916-017-0823-3
关键词: Machine learning 、 Artificial intelligence 、 Decision support system 、 Computer science 、 Rough set 、 Clinical decision support system 、 Data set 、 Missing data 、 Artificial neural network 、 Computational intelligence 、 Data mining 、 Reduct
摘要: As a recent trend, various computational intelligence and machine learning approaches have been used for mining inferences hidden in the large clinical databases to assist clinician strategic decision making. In any target data irrelevant information may be detrimental, causing confusion algorithm degrades prediction outcome. To address this issue, study attempts identify an intelligent approach disease diagnostic procedure using optimal set of attributes instead all present set. proposed Application Specific Intelligent Computing (ASIC) support system, rough based genetic is employed pre-processing phase back propagation neural network applied training testing phase. ASIC has two phases, first handles outliers, noisy data, missing values obtain qualitative generate appropriate attribute reduct sets from input computing centred on relative fitness function measure. The succeeding system involves both classifier selected reducts. model performance evaluated with widely adopted existing classifiers. tested breast cancer, fertility diagnosis heart University California at Irvine (UCI) repository. outperformed attaining accuracy rate 95.33%, 97.61%, 93.04% issue diagnosis.