Post-Earthquake Building Damage State Classification Using Machine Learning: The Case of İzmir

Creative Commons License

Yalçın D., AKTUĞ B.

Turk Deprem Arastirma Dergisi, cilt.8, sa.1, ss.92-103, 2026 (Scopus, TRDizin)

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 8 Sayı: 1
  • Basım Tarihi: 2026
  • Doi Numarası: 10.46464/tdad.1802507
  • Dergi Adı: Turk Deprem Arastirma Dergisi
  • Derginin Tarandığı İndeksler: Scopus, TR DİZİN (ULAKBİM)
  • Sayfa Sayıları: ss.92-103
  • Anahtar Kelimeler: Damage estimation, Geographic Information Systems (GIS), Machine learning, Samos Island (Seferihisar-İzmir) earthquake, Unbalanced dataset
  • Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
  • Ankara Üniversitesi Adresli: Evet

Özet

Highlights * Comparison of damage estimation performance across different machine learning algorithms * Estimation of building damage states using real damage data * Creating a spatial damage estimation map in GIS using machine learning methods Aim To estimate post-earthquake building damage using machine learning and GIS Location Bayraklı and Bornova districts, İzmir, Türkiye Methods Machine learning algorithms (RF, DT, XGB, SVM, ANN) applied to post-earthquake damage data; class imbalance corrected using ROS and SMOTE Results Ensemble-based models provide more reliable post-earthquake damage estimation Post-Earthquake Building Damage State Classification Using Machine Learning: The Case of İzmir