Pengambilan Keputusan Medis Berbasis Algoritma K-nearest Neighbor (Knn) Dalam Klasifikasi Pasien Stroke

Risma Ananta Maulida (Universitas Pamulang)
Suci Anisa Aulia (Universitas Pamulang)
Ridho (Universitas Pamulang)
Satrio Dzulfahmi Yulianto (Universitas Pamulang)
Shania Clara Efendi (Universitas Pamulang)
Maulana Fansyuri (Universitas Pamulang)

Journal of Information Technology and Informatics Engineering
Vol 1 No 1 (2025): Journal of Information Technology and Informatics Engineering (JITIE)

Stroke is a non-communicable disease and one of the leading causes of death and disability worldwide. Early detection of potential stroke risk is crucial to support effective prevention and management efforts. This study aims to develop a stroke risk classification system using the K-Nearest Neighbor (KNN) algorithm implemented through the RapidMiner platform. The dataset analyzed consists of 932 patient records with various medical and demographic attributes. The research process includes data preprocessing, variable transformation, normalization, and splitting the data into training and testing sets. Model evaluation shows an accuracy rate of 82.35%; however, the model has not performed well in identifying stroke cases due to data imbalance. These findings highlight the importance of addressing class imbalance in medical data and the need to consider alternative algorithms to improve detection of minority classes.