基于EM算法的食品污染物分布模型

Models of Contamination Distribution in Food Based on EM Algorithm

食品中污染物分布情况常用对数正态分布来模拟,在使用中面临着小于检测标准的样本数据缺失问题,本文给出了解决该问题的EM算法.为解决EM算法对检测标准过于敏感的问题,进一步建立了由多个分布叠加的改进模型.通过Monte Carlo模拟实验,证明改进模型能更好地模拟食品污染物的分布情况,具有较强的适应性和应用价值.

The contamination distribution in food is simulated with lognormal distribution generally, but it is troubled with the lost data which are less than testing standard in use. An improved model was proposed to resolve the problem that EM algorithm is too sensitive to testing standard. Monte Carlo experiments prove that the improved model can simulate the contamination distribution in food more exactly, and it is more adaptable and has higher application value.