THE GRADED CONSTRUCTION OF NEURAL-NETWORKS

被引：0

作者：

Li Hangping ^{[1
]}

Pan Baochang ^{[1
]}

Wei Yuke ^{[1
]}

机构：

[1] Guangdong Univ Technol, Fac Mat & Energy Sources, Guangzhou 510006, Guangdong, Peoples R China

来源：

ICCSE 2008: PROCEEDINGS OF THE THIRD INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE & EDUCATION: ADVANCED COMPUTER TECHNOLOGY, NEW EDUCATION | 2008年

关键词：

Two layer-grades of NN; Data Fusion-processing; Traditional Chinese Medicine (TCM) diagnosis;

D O I：

暂无

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Such as the Traditional Chinese Medicine (TCM) clinical data, it is a large and complex data-set. There are several different types of data. So the data are divided into two grades. The high grade data belong to a subset, and the low grade data are divided into several subsets. According to the scope that data relate to, the low grade data can be divided into global data and local data. By well using the priori knowledge of data-set about classification and grade, a two layer-grades of neural-network (NN) was built, all layers of NN are divided into two grades, a low-layer-grade sub-NN calculates a low grade sub-data-set and its high layer-grade sub-NN comprehensively processes the high grade sub-data-set and the outputs of the low layer-grade sub-NN. It simplifies the calculation and raises the learning convergent-speed that the two layer-grade NN processes a large and complex data-set. The two layer-grades of NN can be well applied to the TCM categorical identification intelligent calculation provided with complex data relationships.

引用

页码：24 / 26

页数：3