Guraksin, Gur EmreUguz, Harun2020-03-262020-03-2620111349-41981349-418Xhttps://hdl.handle.net/20.500.12395/26229The heart is of crucial significance to human beings. Auscultation with a stethoscope is regarded as one of the pioneer methods used in the diagnosis of heart diseases. However, the fact that auscultation via a stethoscope depends on the skills of the physician's auscultation or his/her experience may lead to some problems in diagnosis. Therefore, the use of an artificial intelligence method in the diagnosis of heart sounds may help the physicians in a clinical environment. In this study, primarily, heart sound signals in numerical format were separated into sub-bands through discrete wavelet transform. Next, the entropy of each sub-band was calculated by using the Shannon entropy algorithm to reduce the dimensionality of the feature vectors with the help of the discrete wavelet transform. The reduced features of three types of heart sound signals were used as input patterns of the least square support vector machines and they were classified by least square support vector machines. In the method used, 96.6% of the classification performance was obtained. The classification performance of the method used was compared with the classification performance of previous studies which were applied to the same data set, and the superiority of the system used was demonstrated.eninfo:eu-repo/semantics/closedAccessLeast squares support vector machineDiscrete wavelet transformShannon entropyHeart soundsArticle71271317144WOS:000297957500037N/A