TATILAH, TITI MEI (2014) Menganalisa Sistem cerdas untuk Deteksi Dini Penyakit Jantung dengan Menggunakan Algoritma Decision Tree & Fuzzy Clustering. Skripsi, Fakultas Ilmu Komputer.
| PDF - Published Version Download (4015b) | Preview |
Abstract
Heart disease is one of the deadly disease in Indonesia. Based on data from the Indonesian Heart Foundation, heart and blood vessel disease in Indonesia reached 26.8 percent and getting closer to the cause of death in patients with an increasingly young age. Early detection of heart disease is needed in order to decrease the death rate from heart attacks. One method that can help in decision making related to diagnostic accuracy of heart disease is the implementation of intelligent systems. At the end of this task, intelligent systems are built through a combination of decision tree and fuzzy clustering algorithms. The algorithm is implemented in matlab applications using the data history of heart disease dataset obtained from UCI Machine Learning Repository. The final result is the creation of intelligent systems that can be complementary tools for doctors in diagnosing heart disease with low levels of disease stage to high. The test result of the peformance system shows that the decision tree classifier peformance gives the AUC ( Area Under ROC ) value amount 0.878. These values are categorized in a good classification category or can be defined that the classifier model can classify the data appropriately.
Item Type: | Article |
---|---|
Subjects: | T Technology > Teknik Informatika > RPL Rekayasa Perangkat Lunak dan Data Universitas Dian Nuswantoro > Fakultas Ilmu Komputer > Teknik Informatika > RPL Rekayasa Perangkat Lunak dan Data |
Divisions: | Library of Congress Subject Areas > T Technology > Teknik Informatika > RPL Rekayasa Perangkat Lunak dan Data Fakultas Ilmu Komputer > Teknik Informatika > RPL Rekayasa Perangkat Lunak dan Data |
Depositing User: | Psi Udinus |
Date Deposited: | 12 Sep 2014 09:42 |
Last Modified: | 22 Nov 2014 06:05 |
URI: | http://eprints.dinus.ac.id/id/eprint/5429 |
Actions (login required)
View Item |