摘要
分析了中医临床记录中症状与症候类别之间的关系,将机器学习中的最大熵原理应用于中医辨症中,建立相应的分类模型,从而观察类别预测的正确性,为中医智能诊断提供初筛和决策支持。同时,将基于最大熵的分类器和基于朴素贝叶斯的分类器进行比较,结果显示,基于最大熵的分类效果胜过朴素贝叶斯分类。这表明将最大熵原理以及算法应用在中医诊断是可行的。
The relationship between clinical medicine symptoms and symptoms categories of records are analyzed. The principle of maximum entropy in machine learning is applied to the TCM syndrome. The corresponding classification model is established to observe the category and forecast category correctness. Intelligent diagnosis for TCM is provided to support the screening and decision support. Compared with the simple Bayesian classifier, the experimental results show that the maximum entropy classification is more effective than Naive Bayes. This suggests that the maximum entropy principle and the algorithm are feasible in the classification of traditional Chinese medicine.
出处
《计算机时代》
2015年第3期50-52,55,共4页
Computer Era
基金
国家自然科学基金青年基金(61202250)
关键词
中医临床记录
最大熵
TCM
中医辨证
the doctor of traditional Chinese medicine clinical record
maximum entropy
TCM
syndrome differentiation
