摘要
数据挖掘技术中的聚类算法是解决客户细分问题的重要算法之一。为解决传统聚类算法在客户细分问题中分类精度较低、收敛速度较慢的问题,着重对比分析传统聚类算法中K-m eans、自组织映射网络和粒子群3种算法的不足,提出融合3种算法优点的混合型聚类算法,该算法利用K-m eans和自组织映射网络对初始聚类中心进行优化,结合粒子群优化和K-m eans优化聚类迭代过程,并在迭代优化过程中设计避免算法因早熟而停滞的机制。针对移动电子商务环境下的餐饮业客户细分问题,建立移动餐饮业客户细分模型,并利用混合型聚类算法、K-m eans、层级自组织映射网络和基于粒子群的K-m eans等4种算法对实际案例进行对比分析。研究结果表明,混合型聚类算法的聚类精度分别比其他3种算法高,同时还具有最快的收敛性能,更适用于客户细分问题。
Clustering algorithms in data mining technology is an important kind of algorithms of soloving customer segmentation problems. To overcome the low accuracy and slow convergence of traditional clustering algorithms in customer segmentation, this paper analyzes deficiencies of traditional cluster algorithms, K-means, SOM and PSO. After that, an improved hybrid clustering algorithm named KSP is proposed, which integrates advantages of K-means, SOM and PSO. The initialization of KSP is optimized by K-means and SOM ; the solving process is carried out by the combination of PSO and K-means with a mechanism of restraining premature stagnancy. Then, a customer segmentation model was established to analyze types of customers in catering industry under mobile electronic commerce environment. Also, an actual case was illustrated to verify the efficiency of the KSP algorithm. The results show that the KSP has the highest accuracy and convergence rate. Thus, it is more suitable for customer segmentation.
出处
《管理科学》
CSSCI
北大核心
2011年第4期54-61,共8页
Journal of Management Science
基金
国家自然科学基金(70890080
70890083)~~
关键词
客户细分
K—means
自组织映射
粒子群优化
混合聚类
customer segmentation
K-means
self-organizing map
particle swarm optimization
hybrid clustering