基于贝叶斯网络的一种牛奶产量预测研究

A Study of the Bayesian Network Applied to Milk Output Prediction

本文运用贝叶斯网络对某农场的牛奶产量进行学习与预测,运用Chi2离散化方法的一种变形进行数据预处理,采用有启发规则指导的、带随机重启的贪心算法搜索网络结构;最后,将贝叶斯网络方法的结果与多元线性回归方法得到的结果进行了比较。

In this paper, the Bayesian network is apphed to learning and predicting the milk output of a farm. In the learning process,a variation of the Chi2 discretization method is used to pre-process the raw data, and the heuristic-rulesguided greedy algorithm with random restart is proposed to search the network structure. Finally, the results yielded by the Bayesian network are con：pared to those by polynomial linear regression.