Triyanna Widiyaningtyas, Ilham Ari Elbaith Zaeni, Putri Yula Wahyuningrum
Coronary Heart Disease (CHD) is one of the diseases that is the highest cause of death in various countries, including Indonesia. There are a number of factors that play an important role in the incidence of CHD, called the CHD risk factor. The technique that can be used to perform early diagnosis in identifying a person at risk of being exposed to CHD is classification. The research aims to implement the method of the synthetic neural network Self-Organizing Map (SOM) for the classification of CHD. SOM as one of the techniques in neural networks that aims to visualize data by reducing the dimensions of the data through the use of self-organizing neural networks. The initial parameters used in this method are learning rate parameters 0.05, minimum learning rate 0.01, and maximum iteration 100. The validation process uses the holdout method. The composition of training data and test data used are 1:1, 1:2, 1:3, and 1:4. The results showed that the most optimal level of accuracy in the data comparison of testing and training is 20%:80%. The values of performance measurement obtained are accuracy 62.5%, precision 60.33%, recall 63.33% and error rate 37.5%. © 2019 IEEE.
State University of Malang, Electrical Engineering Department, Malang, Indonesia
