基于模糊数据挖掘的银行个人客户价值研究

发布时间：2019-02-17 08:49

【摘要】：随着经济全球化的快速发展，信息化管理成为了银行提升竞争力的一项重要手段，而客户价值作为银行的重要组成部分，必然成为了竞争的焦点。由于当前的客户信息软件开发激烈竞争和分析需求多样化的特点，对信息的挖掘带来了极大的挑战，尤其是在客户关系管理中，其客户价值预测标准高、需求变化频繁和限制因素多等客观条件，使客户信息挖掘的难度越来越高。一种基于客户价值指标体系与数据挖掘方法为解决日益增多的信息资源挖掘带来了新的思路，其中模糊数据挖掘方法作为最为一种更精准的挖掘方法以其高度的灵活性得到了极大的关注。本文基于模糊数据挖据的实用性对银行客户价值进行分析与设计。首先，研究银行客户现有问题和特点，分析影响银行客户的因素，分析客户价值的重要性，客户价值指标体系相对于传统方法需求获取的优势。并引入客户细分思想，探求客户价值。然后，依据评价指标的依据与原则，引入专家打分法，构件客户价值指标设计总体框架。分析体系设计步骤，设计客户价值指标体系。最后，分析模糊数据挖掘模式原理与目的，设计数据挖掘模式总体框架，利用模糊数据对客户价值数据的分析，，得到实际划分的结果与预测。针对需求进行模块化设计，并结合X银行客户信息进行实证研究。
[Abstract]:With the rapid development of economic globalization, information management has become an important means for banks to enhance their competitiveness, and customer value, as an important part of banks, must become the focus of competition. Because of the fierce competition of the current customer information software development and the diversification of the analysis demand, it brings great challenge to the information mining, especially in the customer relationship management, its customer value prediction standard is high. It is more and more difficult to mine customer information because of the objective conditions such as frequent change of demand and many restrictive factors. A new method based on customer value index system and data mining has brought new ideas to solve the increasing number of information resources mining. As the most accurate mining method, fuzzy data mining method has been paid great attention to because of its high flexibility. This paper analyzes and designs the customer value of bank based on the practicability of fuzzy data mining. First of all, we study the existing problems and characteristics of bank customers, analyze the factors that affect bank customers, analyze the importance of customer value, and compare the advantages of customer value index system with traditional methods to obtain requirements. And introduce the idea of customer segmentation, explore customer value. Then, according to the basis and principle of evaluation index, the expert scoring method is introduced to design the overall frame of component customer value index. Analyze system design steps and design customer value index system. Finally, the principle and purpose of fuzzy data mining model are analyzed, the general framework of data mining pattern is designed, and the actual partition result and prediction are obtained by using fuzzy data to analyze customer value data. According to the demand of modular design, combined with X bank customer information for empirical research.
【学位授予单位】：哈尔滨理工大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP311.13;F832.3

【参考文献】