Penerapan Algoritma Support Vector Machine (SVM) Pada Pengklasifikasian Penyakit Kejiwaan Skizofrenia (Studi Kasus: RSJ. Radjiman Wediodiningrat, Lawang)

Arya Perdana (Fakultas Ilmu Komputer, Universitas Brawijaya)
Muhammad Tanzil Furqon (Fakultas Ilmu Komputer, Universitas Brawijaya)
Indriati Indriati (Fakultas Ilmu Komputer, Universitas Brawijaya)

Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 2 No 9 (2018): September 2018

Schizophrenia is a disease that attacks a person's psyche, and resulting in behavior with an inappropriate mindset. One of the causes of a person suffering from schizophrenia is stress and also has severe life pressures from various aspects of life. Support Vector Machine (SVM) is an algorithm that can classify types of schizophrenia. The data used in this research is as much as 11 data which is divided into 5 classes. Classes in this study represent five types of diseases in schizophrenia are paranoid, hebefrenik, catatonic, undifferentiated, and simplex. Basically SVM algorithm is a method of linear classification, so that a kernel is used to overcome nonlinear data. In this research is also used One Against All concept to solve multiclass problem. The end result of this research resulted in the highest accuracy of 50.09%, with constant value λ = 1; C = 0,1; γ = 0.1; itermax = 100; ε = 0.01; and also uses polynomial kernels. Tests in this study using K-Fold Cross Validation test, using 11 fold.