OJIT Home

2022 - Volume 5 - Number 1

Time-Series Prediction of Gamma-Ray Counts Using XGB Algorithm

Vincent Mutuku * mulwamutukuu@gmail.com * ORCID: 0000-0002-2469-6482
South Eastern Kenya University, School of Science and Computing, Kitui, KENYA

Joshua Mwema * joshuamke@gmail.com * ORCID: 0000-0002-9002-6794
South Eastern Kenya University, Directorate of Information and Communication Technology, Kitui, KENYA

Mutwiri Joseph * mutwirij3@gmail.com
Tharaka University College, Science Engineering & Technology, Chuka, KENYA

Open Journal for Information Technology, 2022, 5(1), 33-40 * https://doi.org/10.32591/coas.ojit.0501.03033m
Received: 11 March 2022 ▪ Revised: 13 June 2022 ▪ Accepted: 25 July 2022

LICENCE: Creative Commons Attribution 4.0 International License.

ARTICLE (Full Text - PDF)

ABSTRACT:
Radioactivity is spontaneous and thus not easy to predict when it will occur. The average number of decay events in a given interval can lead to accurate projection of the activity of a sample. The possibility of predicting the number of events that will occur in a given time using machine learning has been investigated. The prediction performance of the Extreme gradient boosted (XGB) regression algorithm was tested on gamma-ray counts for K-40, Pb-212 and Pb-214 photo peaks. The accuracy of the prediction over a six-minute duration was observed to improve at higher peak energies. The best performance was obtained at 1460keV photopeak energy of K-40 while the least is at 239keV peak energy of Pb-212. This could be attributed to higher number of data points at higher peak energies which are broad for NaITi detector hence the model had more features to learn from. High R-squared values in the order of 0.99 and 0.97 for K-40 and Pb-212 peaks respectively suggest model overfitting which is attributed to the small number of detector channels. Although radioactive events are spontaneous in nature and not easy to predict when they will occur, it has been established that the average number of counts during a given period of time can be modelled using the XGB algorithm. A similar study with a NaITi gamma detector of high channel numbers and modelling with other machine learning algorithms would be important to compare the findings of the current study.

KEY WORDS: radioactivity, extreme gradient boost, regression, Gamma-rays, photo-peaks, NaITi.

CORRESPONDING AUTHOR:
Vincent Mulwa, South Eastern Kenya University, School of Science and Computing, Kitui, KENYA. E-mail: mulwamutukuu@gmail.com.

REFERENCES:

Akkurt, I.,  Gunoglu, K., & Arda, S. (2014). Detection efficiency of NaI(Tl) Detector in 511–1332 keV energy range,” Science and Technology of Nuclear Installations, vol. 2014, 2014.

Althnian, A., AlSaeed, D., Al-Baity, H., Samha, A., Dris, A., Alzakari, N., Abou, A., Elwafa, & Kurdi, H. (2021). Impact of Dataset Size on Classification Performance: An empirical evaluation in the medical domain. Appl. Sci.,11, p. 796.

Asaduzzaman, K., Mannan, F., Khandaker, M., Farook, M., Elkezza, A., & Amin, Y. (2015). Assessment of natural radioactivity levels and potential radiological risks of common building materials used in Bangladeshi dwellings. PLOS ONE,10.

Aslam, M., Gul, R., Ara, T., & Hussain, M. (2012). Assessment of radiological hazards of naturally occurring radioactive materials in cement industry. Radiation Protection Dosimetry, 3(151), 483-488.

Connell, C., & Pike, G. (2005). Encyclopedia of Analytical Science (Second Edition).

Ebbing, D., & Wentworth (1995). In Experiments in Introductory Chemistry, p. 221.

Hossain, I., Sharip, N., & Viswanathan, S. (2012). Efficiency and resolution of HPGe and NaI(Tl) detectors using gamma-ray spectroscopy. Scientific Research and Essays, 7(1), 86-89.

Keeble, J., & Rios, A. (2020). Machine learning the deuteron. Physics Letters B.

Klaus, G., & John, C. (1995). Neural networks that learn to predict probabilities: Global models of nuclear stability and decay. Neural Networks, 8(2), 291-311.

James, P., & Christine, C. (2015). Compton scattering of Cs-137 Gamma rays. Department of Physics and Astronomy, The University of Tennessee, 401 Nielsen Physics Building, Knoxville.

Niu, M., Liang, Z., Sun, H., Long, W., & Niu, F. (2019). Predictions of nuclear β-decay half-lives with machine learning and their impact on r-process nucleosynthesis, Physics Review C.

NRCC (1999). Natural radioactivity and radiation. National Academies Press (US), Washington (DC).

S., P., Freitas, A., & John, C. (2019). Experiments in machine learning of alpha-decay half-lives. Nuclear Theory.

Saxena, G., Sharma, K., & Prafulla, S. (2021). Modified empirical formulas and machine learning for α-decay systematics. Journal of Physics G: Nuclear and Particle Physics, 48(5).

Sharma, N., Singh, J., Esakki, C. A., & Tripath, R. (2016). A study of the natural radioactivity and radon exhalation rate in some cements used in India and its radiological significance. Journal of Radiation Research and Applied Sciences, 9(1), 47-56.