Evidence record · UDINUS · SDGs
- Unit
- Rektorat UDINUS
- Period
- 2024
- Goals
- Record
- 5df59b6e-2d92-42a6-b489-b60707226b2e
- Published
- 2026-10-30
Universitas Dian Nuswantoro conducted research on heart disease classification using the Deep Neural Network algorithm and the SMOTE technique to overcome data imbalance. The activity involved Cahyani, Ailsa Nurina; Zeniarja, Junta; Winarno, Sri; Putri, Rusyda Tsaniya Eka; and Maulani, Ahmad Alaik as participants. The research addressed heart disease, which the source describes as the leading cause of premature death worldwide. According to the WHO statement included in the source, heart disease causes about 30% of the total 58 million deaths and mainly occurs in individuals who are in their productive age.
The activity examined the application of a Deep Neural Network algorithm for classifying heart disease and the use of SMOTE as a data-balancing technique. The source places this work among studies using various algorithms, methods, and data-balancing techniques to anticipate heart disease and obtain better accuracy results. The reported validation accuracy was 90%. The reported precision evaluation was 0.85, the recall was 0.92, and the F1-score was 0.88. Based on these results, the Deep Neural Network algorithm after SMOTE was reported as superior to the model without SMOTE. This UDINUS research relates to the health focus identified in RAW_GOALS through its examination of heart disease classification and data balancing. The available evidence covers the research objective, the algorithm and technique used, the evaluation results, and the comparison between the model after SMOTE and the model without SMOTE.