Prediction of Gamma Radiation Shielding Properties Using Machine Learning Methods

Creative Commons License

İçkecan D., Aydın E. E., Türkan M. N.

Gamma Rays - Science, Technology, and Applications, Associate Prof. Muhammad Zubair, Editör, IntechOpen, London, ss.1-150, 2026

  • Yayın Türü: Kitapta Bölüm / Araştırma Kitabı
  • Basım Tarihi: 2026
  • Yayınevi: IntechOpen
  • Basıldığı Şehir: London
  • Sayfa Sayıları: ss.1-150
  • Editörler: Associate Prof. Muhammad Zubair, Editör
  • İstanbul Ticaret Üniversitesi Adresli: Evet

Özet

This chapter explores the increasing need for effective radiation shielding materials in various fields, including medicine, industry, agriculture, and nuclear technologies. As gamma radiation poses significant risks, understanding the linear attenuation coefficient (LAC) of materials is essential for designing safe and efficient systems. Traditionally, LAC values are obtained through experimental measurements or Monte Carlo simulations, both of which can be time-consuming and costly. In this study, machine learning models—Support Vector Regression, Kernel Ridge Regression, and k-Nearest Neighbors—are trained using LAC data from the NIST XCOM database covering 0.01–10 MeV. The models are then tested on Praseodymium Hexaboride (PrB₆) to evaluate their predictive capability. Findings show that machine learning provides a fast, accurate, and cost-effective alternative for estimating gamma attenuation properties, offering valuable support for the development and optimization of radiation-related technologies.

This chapter explores the increasing need for effective radiation shielding materials in various fields, including medicine, industry, agriculture, and nuclear technologies. As gamma radiation poses significant risks, understanding the linear attenuation coefficient (LAC) of materials is essential for designing safe and efficient systems. Traditionally, LAC values are obtained through experimental measurements or Monte Carlo simulations, both of which can be time-consuming and costly. In this study, machine learning models—Support Vector Regression, Kernel Ridge Regression, and k-Nearest Neighbors—are trained using LAC data from the NIST XCOM database covering 0.01–10 MeV. The models are then tested on Praseodymium Hexaboride (PrB₆) to evaluate their predictive capability. Findings show that machine learning provides a fast, accurate, and cost-effective alternative for estimating gamma attenuation properties, offering valuable support for the development and optimization of radiation-related technologies.