Estimation of Cosmic Radiation Dose Received on Flight by Means of Different Machine Learning Methods and CARI-7A Calculations
Romanian Journal Of Physics, cilt.67, sa.1-2, ss.1-18, 2022 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 67 Sayı: 1-2
- Basım Tarihi: 2022
- Dergi Adı: Romanian Journal Of Physics
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Chemical Abstracts Core, zbMATH
- Sayfa Sayıları: ss.1-18
- İstanbul Üniversitesi-Cerrahpaşa Adresli: Evet
Özet
Sixteen machine learning methods including K-Nearest-Neighbor,
Random Forest, Additive Regression, Linear Regression and M5P etc. were used to
estimate the cosmic radiation dose for different international flights related with
Istanbul and Ankara Airports in Turkey. Latitude, longitude and depth were used as
inputs to the developed models, and the output variable is the dose rate. In order to
evaluate accuracy of the developed models, five statistical indicators; correlation
coefficient (R), mean absolute error (MAE), root mean squared error (RMSE), relative
absolute error (RAE) and root relative squared error (RRSE) were compared. The
results showed that K-Nearest-Neighbor (k-NN) model approach provides a high
performance and lower error to predict the dose rate. Besides machine learning
methods, the dose values in the flights were calculated with the CARI-7A software
and the results obtained with both methods were compared and seen that they were in
good agreement.