Performance Comparison of Classification Algorithms Used in Stroke Prediction Inme Tahmininde Kullanilan Siniflandirma Algoritmalarinin Performans Karsilastirmasi

Unsal B., Kaplan K., Küçüksarı Ö.

33rd IEEE Conference on Signal Processing and Communications Applications, SIU 2025, İstanbul, Türkiye, 25 - 28 Haziran 2025, (Tam Metin Bildiri) identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/siu66497.2025.11112188
  • Basıldığı Şehir: İstanbul
  • Basıldığı Ülke: Türkiye
  • Anahtar Kelimeler: Classification, Hyperparameter, Imbalanced Dataset, Machine Learning, Stroke Prediction
  • Kocaeli Üniversitesi Adresli: Evet

Özet

Stroke is a life-threatening condition that occurs as a result of interrupted blood flow to the brain. As with many diseases, early diagnosis and intervention play a critical role in preventing stroke cases. This study aims to predict strokes using different machine learning algorithms. In this context, Logistic Regression (LR), Random Forest (RF), Decision Tree (DT), and Gaussian Naive Bayes (NB) models were applied, and the accuracy, F1 score, recall, and precision performances of the models were compared. Due to the imbalanced distribution of the dataset used in the study, the effects of class imbalance on model performances were examined. At the end of the study, the models providing the highest performance were identified. Additionally, the impact of data balancing on the models' prediction results has been presented. By performing hyperparameter optimization, it was aimed to increase the prediction accuracy of the models.