摘要
提出用粗糙集理论对中药样本进行属性约简,接着基于单核函数的局限性,构造混合核SVM中药功效分类器,预测骨性关节炎中药复方中药物的功效.经中医理论知识验证,分类精度高且合理.结合了粗糙集理论消除冗余属性,减少了原始支持向量机的运算量,提高了模型的识别性能.
The paper used the theory of rough set to reduce the attributes of the traditional Chinese medicine samples,and then build a classifier for Chinese medicine efficacy based on mixed kernel SVM to predict the efficacy of osteoarthritis compound drugs,while single kernels have defects.Verified by the traditional Chinese medicine theory,the result of the classification is accurate and reasonable.In the paper,under the help of rough set to remove redundant attributes is not only to reduce the calculation of original SVM,but also to improve the recognition performance of the model.
出处
《福州大学学报(自然科学版)》
CAS
CSCD
北大核心
2013年第3期311-316,共6页
Journal of Fuzhou University(Natural Science Edition)
基金
福建省自然科学基金资助项目(2009J01282
2012J01261)
国家自然科学基金资助项目(61104041
61201397)
关键词
骨性关节炎
中药
粗糙集
属性约简
SVM
osteoarthritis
Chinese medicine
rough set
attribute reduction
SVM