摘要
针对聚类算法在金融领域广泛应用的实际情况,基于银行客户数据集,对DBSCAN,K-means和X-means 3种聚类算法在执行效率、可扩展性、异常点检测能力等方面进行对比分析,并提出将X-means算法应用于银行业客户细分。利用X-means算法建立了一套银行客户细分模型,为银行决策者提供科学的决策支持。
This paper considers cluster analysis, which is the most often applied in commerce area and discusses three algorithms: DBSCAN, K-means and X-means based on the bank customer standard dataset. It compares algorithms concerning their effectiveness, scalability, outliers detection ability, and builds some bank customer models with X-means, which provides more powerful suggestions for bank decision-making man.
出处
《计算机工程》
CAS
CSCD
北大核心
2008年第24期37-39,共3页
Computer Engineering
基金
中国科学院院长基金资助项目(A050414)