Psychological Pharmacokinetics Model Based on Bayes Network with Optimal of Kernel Density Estimation of Prior Distribution

Neural network is widely used in pharmacokinetics and psychology area. The psychology of patients can affect the pharmacokinetics. In order to get the prior distribution of psychological pharmacokinetics parameters, we introduce Bayes method to describe that, which is more scientific. The key point of inductive-learning in Bayes network is the estimation of prior distribution. This paper adopts general naive Bayes to handle the psychology pharmacokinetics parameter data, and proposes a kind of kernel function constructed by orthogonal polynomials, which is used to estimate the density function of prior distribution in Bayes network. Then, the paper makes further researches in the optimality of the kernel estimation of density and derivatives. When the sample is fixed, the estimators can keep continuity and smoothness, and when the sample size tends to infinity, the estimators can keep good convergence rates.