摘要
模糊k-prototypes算法是当前聚类分析中最有效算法之一.简述了模糊k-prototypes算法的发展进程和主要性质;并在此基础上,指出它在处理数值型和分类型混合数据时的不足,进而提出一种改进算法;最后,将算法应用到英语借词之中,给出计算结果.结果表明,改进算法具有较好的稳定性和较高的精确度.
Fuzzy k-prototypes algorithm is one of the most efficient algorithms in clustering at present. This paper describes the developing process and the main properties for fuzzy k-prototypes algorithm, and points out its disadvantages in dealing with the mixed numeric and categorical valued data; furthermore, an improved algorithm is presented; finally, the algorithm is applied to Chinese loanwords in English. The computing results show the good stability and high accuracy.
出处
《大连理工大学学报》
EI
CAS
CSCD
北大核心
2003年第6期849-852,共4页
Journal of Dalian University of Technology