Predicting cardiovascular disease with hybrid learning

The age of getting heart and vascular diseases risk, which decrease the quality of life in our country and world, cause lots of deaths and has a high cost of treatment, decrease day by day. The role of early diagnosis in the treatment of disease is really important. There are many works related this field nowadays. In recent days, with the development of technology, machines imitate the human learning and make predictions. With this method, which called Artificial Intelligence, machines are thought to facilitate the daily life. Artificial neural networks, which are the lower branches of artificial intelligence, are used for early diagnosis of diseases. With these methods used in many studies in the literature, the factors causing diseases can be taken as input and the disease result can be estimated. Therefore, early diagnosis of those who are at risk and treatment could be possible. In this thesis study, the aim is to determine the disease in early diagnosis by estimating the treatment of cardiovascular diseases. This study used artificial neural network and k-means algorithm. Therefore, there exist a hybrid system. Two different models of artificial neural network and hybrid system have been developed in order to determine the model that gives the best result in the prediction of disease of cardiology data. To determine the best result model, the mean squared error, root mean squared error and average absolute percent error used in the model success criterion were used.