基于Apriori与User-CF的银行客户挖掘及个性化推荐

Bank Customer Mining and Personalized Recommendation Based on Apriori and User⁃CF

大数据技术的发展向银行的数据挖掘能力提出新要求,因此加强银行智能化服务,提升银行的金融服务能力对提升银行的客户满意度具有现实意义。采用Apriori关联规则算法,分析办理业务的各类型客户间的关联规则,有利于挖掘能够发展转化为个贷客户的潜在新用户;为此,本研究分别构建余弦相似度与Pearson相似度的协同过滤算法,预测客户对未知类型产品的偏好程度,根据相关评分将相应产品推荐给客户。结果表明,通过Apriori关联规则算法与协同过滤算法,能够有效提高数据的分析管理能力,助力银行挖掘新客户,深度培育基础客户,为客户提供更加个性化、人性化的服务系统,加强银行智能化服务,从而提升银行的金融服务能力。

The development of big data technology puts forward new requirements for banks’data mining capabilities.Strengthening banks’intelligent services and improving banks’financial service capabilities are of practical significance for improving banks’customer satisfaction.The study uses Apriori association rule algorithm to analyze the association rules among various types of customers,which is conducive to mining potential new applications that can be developed into individual loan customers.Therefore,the study constructs the collaborative filtering algorithms of cosine similarity and Pearson similarity respectively to predict the degree of customer preference for unknown types of products,and recommends corresponding products to users according to the relevant scores.The results show that the Apriori association rule algorithm and collaborative filtering algorithm can effectively improve the data analysis and management ability,help banks to dig new customers,deeply cultivate basic customers,provide customers with a more personalized and humanized recommendation system,and strengthen the bank’s intelligent service,so as to improve the bank’s financial service ability.