IMPLEMENTASI ALGORITMA RANDOM FOREST DALAM PREDIKSI STROKE DENGAN PENGGUNAAN SYNTHETIC MINORITY OVER-SAMPLING TECHNIQUE

• Main Author: Johanes Gerald, (Author)
• Format: Book
• Published: 2024-01-15.
• Subjects: Thesis; NonPeerReviewed
• Online Access: Link Metadata
• Notes: http://repository.upnvj.ac.id/27829/1/ABSTRAK.pdf; http://repository.upnvj.ac.id/27829/24/AWAL.pdf; http://repository.upnvj.ac.id/27829/4/BAB%201.pdf; http://repository.upnvj.ac.id/27829/5/BAB%202.pdf; http://repository.upnvj.ac.id/27829/6/BAB%203.pdf; http://repository.upnvj.ac.id/27829/7/BAB%204.pdf; http://repository.upnvj.ac.id/27829/8/BAB%205.pdf; http://repository.upnvj.ac.id/27829/9/DAFTAR%20PUSTAKA.pdf; http://repository.upnvj.ac.id/27829/12/RIWAYAT%20HIDUP.pdf; http://repository.upnvj.ac.id/27829/11/LAMPIRAN.pdf; http://repository.upnvj.ac.id/27829/10/HASIL%20PLAGIARISME.pdf; http://repository.upnvj.ac.id/27829/2/ARTIKEL%20KI.pdf
• Keywords: Q Science (General)

Stroke is one of the significant global health issues, being a leading cause of disability and death worldwide. In Indonesia, stroke has emerged as one of the most fatal diseases. According to the 2020 Indonesia health profile data, stroke ranks third with a total of 1,789,261 reported cases. The aim of this research is to identify patients at high risk of stroke. The algorithm employed is the Random Forest Classifier utilizing the Synthetic Minority Oversampling Technique to balance the class data. In the Random Forest method using the Synthetic Minority Oversampling Technique, the results showed an accuracy of 95.61%, precision of 93.66%, recall of 97.85%, and an f1-score of 95.71%. Meanwhile, for the Random Forest model, the accuracy was 90.15%, precision was 90.5%, recall was 90.15%, and the f1-score was 90.32%. Due to class imbalance, using the Random Forest algorithm alone is not suitable without resampling. Therefore, Random Forest - SMOTE can be utilized as one of the algorithms to predict strokes.