本文已被：浏览 1205次 下载 1912次
Received:April 22, 2014 Revised:November 28, 2014
Received:April 22, 2014 Revised:November 28, 2014
中文摘要: 对作者之前设计的求语言特征主成分的“距离法”做了深度解析, 指出了“距离法”在应用过程中, 切分成两个集合的“断点”会影响计算结果的精度. 借助聚类分析的思想重新设计了语言特征主成分算法——“聚类法”来弥补“距离法”的缺陷. 最后指出两个方法的缺陷可以相互弥补, 在实际应用中应结合使用.
Abstract:This paper analyzes the “distance method” that is an algorithm of language features and points out the questions of “distance method”: “Breakpoint” that be used cut into two sets will affect the accuracy of calculation. It designs a new algorithm named “clustering method” with the idea of clustering analysis. The defects of two methods can complement each other and should be used together in practical application.
文章编号： 中图分类号： 文献标志码：
基金项目:教育部哲学社会科学研究重大课题攻关项目(09JZDH007)
| Author Name | Affiliation |
| SHAO Ze-Guo | Institute of Language, Shanghai Normal University, Shanghai 200234, China Shanghai Instrumentation Colledge, Shanghai 200940, China |
| Author Name | Affiliation |
| SHAO Ze-Guo | Institute of Language, Shanghai Normal University, Shanghai 200234, China Shanghai Instrumentation Colledge, Shanghai 200940, China |
引用文本：
邵泽国.特征主成分算法再探.计算机系统应用,2015,24(7):137-141
SHAO Ze-Guo.Restudy of Features Principal Component Algorithm.COMPUTER SYSTEMS APPLICATIONS,2015,24(7):137-141
邵泽国.特征主成分算法再探.计算机系统应用,2015,24(7):137-141
SHAO Ze-Guo.Restudy of Features Principal Component Algorithm.COMPUTER SYSTEMS APPLICATIONS,2015,24(7):137-141
