摘要
根据某通信运营商经营分析系统的要求,改进并实现了一种新的聚类算法,对该运营商的客户概况进行分析。在运算过程中,滤去与其它类较远且类内数据点较少的噪声数据,消除噪声数据对整体聚类的影响,在一定程度上减少了人为因素的影响。从该系统的实际运行情况看,效率较高,结果可信,达到了客户的要求。
he new clustering arithmetic is ameliorated and implemented, which in on the foundation of a communicate proprietor in this paper. The new clustering arithmetic can wipe offthe useless datas on the course of operation and expedit the velocity of clustering. On the instance of practice circulating, the new clustering arithmetic has a higher efficiency and.a trust result, It satisfies the request of clients.
出处
《长春理工大学学报(自然科学版)》
2007年第4期131-134,共4页
Journal of Changchun University of Science and Technology(Natural Science Edition)
关键词
客户分析
通讯
聚类算法
计算机应用
client analysis
communication
clustering arithmetic
computer application