Implementation of Support Vector Machine and Multilayer Perceptron Algorithms for Patient Diagnosis Based on Patient Profile and Complaints at Jatibening Public Health Center

Romanda Ilham (Darma Persada University)
Afri Yudha (Darma Persada University)

Journal Technology Information and Data Analytic
Vol 2 No 2 (2025)

Community health centers (Puskesmas) are primary healthcare institutions that play a crucial role in providing services to the community, especially in areas with limited access. However, the disease identification process at the Jatibening Community Health Center still uses traditional methods that are time-consuming and potentially biased. This study aims to create a disease prediction system for patients using the Support Vector Machine (SVM) and Multilayer Perceptron (MLP) machine learning algorithms that utilize data from patient profiles and complaints. The methods used in this study include collecting information from patient medical records, data processing, training SVM and MLP models, and assessing the model's accuracy level. Test results show that the MLP algorithm achieves 100% accuracy, while the SVM also demonstrates 100% accuracy in predicting the likelihood of a patient's disease based on factors such as age, gender, and chief complaint. Thus, the use of machine learning algorithms on patient data at the Jatibening Community Health Center can accelerate the initial diagnosis process and support more efficient medical decision-making