Application and improvement of continuous monitoring methods for artificial radionuclides based on Bayesian statistics.

This paper delves into the Bayesian statistics applications of three preeminent models, Poisson distribution, Gaussian distribution, and Binomial distribution, in the continuous surveillance of artificial radionuclides. It introduces a slide-window method to accelerate the updating of the prior distribution of model parameters and compares the performances of three models before and after utilizing this method. Comparisons among the three models are made before and after using the slide-window. Experimental results demonstrate a marked enhancement in the performances of all models. [ABSTRACT FROM AUTHOR]