Comparison of Random Forest Algorithm Accuracy With Xgboost Using Hyperparameters
Year: 2024
Type: article
Abstract: Diabetes is one of the most dangerous diseases in the world and many people do not realize that they have diabetes in them. So many factors affect the occurrence of diabetes such as pregnancies, glucose, blood pressure, skinthickness, insulin, BMI, diabetes pedigree function, and age. so diabetes threatens silently and will appear suddenly. Therefore, this study will make a diabetes prediction using Random Forest and XGBoost algorithms. The model will be evaluated with accuracy, F1-Score, recall, and precision. for randomization or random state will use random states 0 and 45. The results obtained from the comparison of these two algorithms are the highest accuracy of the random forest algorithm has a value of 88,98% while the highest accuracy of XGBoost gets an accuracy value of 87,00% at random state 45 and data division 90/10, while random state 0 random forest has the highest accuracy value also with a value of 78,43% with data division 90/10 while XGBoost gets the highest accuracy value of 76,47% at data division 90/10. It can be concluded that random forest is better at predicting diabetes data than the XGBoost algorithm.
Source: Proxies Jurnal Informatika
Authors: Kevin Stefanus, Hironimus Leong
Institution: Soegijapranata Catholic University
Language: English
Subfield: Health Information Management
Field: Health Professions
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W4402005359