期刊文献+

模拟退火K均值聚类算法及其应用研究 被引量:12

Application Research of Simulated Annealing K-Means clustering algorithm
下载PDF
导出
摘要 针对CRM客户分类,提出模拟退火算法与K均值算法相结合的聚类算法。利用模拟退火算法全局寻优能力改变k均值算法易陷入局部极值的缺点。经标准数据集检验,证明算法有效。根据烟草商业企业业务数据和卷烟营销特点分析,设计客户分类评价指标模型。将算法应用于烟草商业企业CRM客户分类,分类结果符合卷烟营销特点,从实用角度验证算法有效。根据客户分类设计了差异化CRM营销策略。 Propose a simulated annealing K-means (SAKM) algorithm for CRM customer clustering, which use the global optimize a- bility of SA to remedy the local extremum shortcoming of KM. The standard data computational results indicate that it is better than the K-means algorithm. A clustering attributes model had been designed which come from the analyzing of tobacco commercial enter- prise' sale data and characteristic. Testify the SAKM in tobacco commercial enterprise CRM customer clustering, the result validate SAKM is in effect, and then propose a personal CRM strategy.
出处 《微计算机信息》 北大核心 2008年第21期182-184,共3页 Control & Automation
关键词 数据挖掘 聚类 模拟退火算法 K均值算法 评价指标模型 烟草商业企业 data mining clustering simulated annealing k-means attributes model tobacco commercial enterprise
  • 引文网络
  • 相关文献

参考文献4

  • 1刘秋菊,王仲英,刘素华.基于遗传模拟退火算法的模糊聚类方法[J].微计算机信息,2006,22(02Z):270-272. 被引量:18
  • 2Shehroz S. Khan and Amir Ahmad, Cluster center initialization algorithm for K-means clustering[J].Pattern Recognition Letters, 25 (2004) 1293 - 1302.
  • 3Ayad H, Kamel M. Topic discovery from text using aggregation of different clustering methods [A].In: Cohen R, Spencer Bed. Advances in artificial intelligence: 15th conference of the Canadian society for computational studies of intelligence [C].Calgary, 2002.161-175.
  • 4Liu D R, Shih Y Y. Integrating AHP and data mining for product recommendation based on customer lifetime value [J]. Information & Management, 2005.42 (7): 387 -400.

二级参考文献3

  • 1姚冰花.基于PROFIBUS-DP总线的Danfoss变频器应用[J].微计算机信息,2005,21(2):78-79. 被引量:11
  • 2康立山 谢云.非数值并行计算-模拟退火算法[M].北京:科学出版社,1995..
  • 3Doloca,adrian.Feature Selection for Texture Analysis Using Genetic Algorithms[J].International Journal of Computer Mathematics,2000;74:279-292.

共引文献17

同被引文献81

引证文献12

二级引证文献31

;
使用帮助 返回顶部