基于数据挖掘技术的商业银行个人信用评分模型研究

发布时间：2018-01-05 05:10

本文关键词：基于数据挖掘技术的商业银行个人信用评分模型研究　出处：《湖南大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着我国经济的持续快速发展和人们收入水平的不断提高，人们的消费观念也随之发生了巨大的变化，个人信贷消费已经成为日常消费中的重要组成部分。因此，建立个人信用等级评估与预测体系既是我国银行业发展的需要，也是促进个人信贷业务发展的有效措施。这就要求我国构建个人信用体系，个人信用评分是个人信用体系建设的关键，通过对个人进行科学的信用评分，可以准确的了解个人的信用状况，正确识别个人信用风险，节约交易成本，提高交易效率，促进信用资源的优化配置。随着计算机技术、数据库技术的快速发展和广泛应用，各行业中积累的数据越来越多，金融行业尤其如此。科学数据的大量积累和各种数据库的广泛使用，人们逐步认识到海量数据的利用十分困难、效率低下，而且很难从中获得有价值的指导性意见。在这种情况下，数据挖掘技术应运而生。数据挖掘是一种新的商业信息处理技术，其主要功能是对商业数据库中的大量业务数据进行抽取、转换、分析和其他模型化处理，从中提取辅助商业决策的关键性数据。本文将数据挖掘技术运用于商业银行的个人信用评分。本文在总结国内外个人信用评分发展现状的基础上，从数据挖掘角度对当前国内商业银行成熟的个人信用评分模型进行介绍和分析。借助于模糊层次分析法进行个人信用评价指标体系的建立，有助于综合考虑个人风险、个人信用状况以及不同指标等情况，避免单一指标考核所存在的问题，，且操作简单有助于规避潜在的信用风险。与此同时，针对模糊矩阵中约束条件的缺陷，本文通过引入总体均值检验理论对模糊层次分析法进行了改进。最后重点分析了基于层次分析法的个人信用评分体系，借鉴国内外的研究成果构建了个人信用评分指标体系，并运用模糊层次分析法确定了个人信用评分指标体系中各指标的权重，将结果应用于实例。本文的研究从理论上丰富了商业银行进行个人信用评分的方法，为商业银行选择最合适的数据挖掘方法进行个人信用评分提供了决策参考，有利于商业银行控制经营风险和提高自身竞争力。
[Abstract]:With China's sustained and rapid economic development and continuous improvement in people's income level, people's consumption concept has also undergone tremendous changes, the personal credit consumption has become an important part in daily consumption. Therefore, to establish the development of personal credit evaluation and prediction system is China's banking industry, but also to promote the effective measures for the development of personal credit business. This requires the construction of China's personal credit system, personal credit scoring is the key to the construction of personal credit system, through scientific credit score for the individual, can accurately understand the personal credit, the correct recognition of personal credit risk, reduce transaction costs, improve transaction efficiency, promote optimal allocation of credit resources. With the development of computer technology, the rapid development and wide application of the database technology, the data accumulated in the industry more and more, especially in the financial industry So it is widely used. The accumulation of scientific data and databases, people gradually realize that the utilization of massive data is very difficult, inefficient, and difficult to obtain guidance value from it. In this case, the data mining technology. Data mining is a new business information processing technology. Its main function is to a large number of commercial databases in the business of data extraction, transformation, analysis and handling of other models, key data extraction from business decisions.
This paper will use data mining technology in the commercial bank personal credit score. Based on summarizing the domestic and foreign development status of credit rating in this paper on the mining perspective on the introduction and analysis of the current domestic commercial banks mature individual credit scoring model from the data. Using fuzzy AHP to establish evaluation index system of personal credit. Contribute to the comprehensive consideration of personal risk, personal credit conditions and different indicators, to avoid the existing single index evaluation of the problem, and the operation is simple and helps to avoid the potential credit risk. At the same time, aiming at the defects of fuzzy constraints in the matrix, through the introduction of the overall mean test theory of fuzzy analytic hierarchy process improved. Finally the article analyses the AHP of personal credit scoring system based on the reference of the research results at home and abroad to build a personal letter With the evaluation index system, and uses fuzzy AHP to determine the weight of each index in the personal credit scoring system, the results can be applied in examples. This study has enriched the theory of commercial bank personal credit scoring methods, choose the most suitable data mining methods for personal credit scoring provides decision-making reference for commercial bank, commercial bank can control the risk and improve their competitiveness.

【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP311.13;F832.4

【参考文献】