Penggunaan Heaviside Activation Function pada Regresi Linear untuk Klasifikasi Diabetes

Felix Indra Kurniadi (Tanri Abeng University)
Vinnia Kemala Putri

Jurnal ULTIMATICS
Vol 10 No 1 (2018): Ultimatics : Jurnal Teknik Informatika

Diabetes is one of the diseases that rapidly increase in the world. One of the most used dataset for diabetes is Pima indian dataset. Pima indian have 8 features such as pregnancies, glucose, blood pressure, insulin, BMI, diabetes pedigree function and age. In this research we are comparing between Linear Regression using Heaviside Activation Function and Logistic Regression. Logistic regression gives better result compare linear regression using Heaviside Activation Function. Index Terms—Diabetes, Regresi, Heaviside Activation Function, Logistic Regression