Penggunaan Principal Component Analysis Dan Pohon Keputusan Untuk Mendeteksi Penyakit Jantung Koroner (original) (raw)

2012

Abstract

There are many ways that can be done to make the diagnosis of a disease like by seeing physical symptoms suffered by patients and utilizing technology to see inside the patient's body. Another way that can be done to detect a disease is using calculations. There are many methods or algorithms that can be used to develop disease detection system for example, artificial neural network method, Naive Bayes, decision trees, rough set theory, principal component analysis (PCA), nearest neighbor, and other. This study uses data for coronary heart disease to learn and discover patterns that can be used to detect coronary heart disease. Principal component analysis method is used to reduce existing variable in the data and result in principal component. Decision tree method is used to create the rules which are used for coronary heart disease diagnosis. From 14 variables that exist in the research data, only nine variables in the research data are considered to have significant effect in performing coronary heart disease diagnosis. Further analysis of the nine variables is done using decission tree methods and results in 25 rules for coronary heart disease detection. The use of two algorithms have accuracy of 75.42%.

Galih Hendro hasn't uploaded this paper.

Let Galih know you want this paper to be uploaded.

Ask for this paper to be uploaded.