《基于支持向量机的企业赊销信用风险管理系统研究》由会员分享，可在线阅读，更多相关《基于支持向量机的企业赊销信用风险管理系统研究（66页珍藏版）》请在金锄头文库上搜索。
1、成都理工大学硕士学位论文基于支持向量机的企业赊销信用风险管理系统研究姓名：刘翔申请学位级别：硕士专业：管理科学与工程指导教师：刘红军;党庆忠20090501摘 要 I 基于支持向量机的企业赊销信用风险管理系统研究基于支持向量机的企业赊销信用风险管理系统研究 作者简介：姓名 刘翔，性别 男，1983 年 12 月生，师从成都理工大学刘红 军教授，2009 年 6 月毕业于成都理工大学管理科学与工程专业，获得管理学硕 士学位。 摘 摘 要要 信用是市场经济的基础，是市场经济正常运行的前提。信用风险无处不在， 迫切需要我们尽快认识信用和信用经济的本质和特点， 提高识别和规避信用风险 的能力，增强企业
2、信用管理水平，适应信用经济发展的需要。 赊销风险属于信用风险的一种。 赊销风险即赊购商品的客户可能无法全额支 付账款，从而导致企业面临着重要的金融风险。在当前的企业赊销信用风险管理 中存在着很多问题，需要我们去改善，去解决。 本论文结合当前企业赊销的信用管理实践，对目前信用管理研究成果进行改 进， 建立赊销信用风险管理框架， 在新的框架下， 根据赊销信用管理的业务流程， 发现赊销信用风险管理的模型，包括筛选客户模型和客户信用风险动态监察模 型。利用 SVM 技术建立数学模型，用先进的分类预测方法建立更为有效的信用 风险评估模型。对企业赊销信用管理系统进行研究，并通过某机械制造业企业的 信用管理
3、实践，对系统的应用进行分析。 本论文的主要研究内容如下： （1）分析了当前我国企业赊销信用风险产生的原因以及信用风险管理中存 在的问题，通过分析，我们得到了企业赊销信用管理的几点启示。即建立独立的 信用管理部门、进行业务流程再造、以及采用先进信息技术建立赊销风险管理系 统。 （2）介绍了当前国内外普遍的企业风险管理框架，根据前面分析的企业赊 销信用管理的启示，建立了企业赊销信用风险管理框架。包括建立企业信用管理 部门，制订企业赊销风险管理政策，赊销风险管理流程分析，以及根据管理流程 得到风险管理的模型结构。 （3）介绍了当前信用评估方法的研究情况，基于多分类 SVM 模型构建赊 销信用风险评估
4、模型。将赊销信用风险评估指标分为定量指标和定性指标，定量 指标主要是客户的财务指标，定性指标主要为客户的行业环境和内部管理因素； 对客户评估数据进行预处理，定量数据采取归一化处理，定性数据利用模糊定量 化方法进行量化。 （4）对企业赊销风险管理系统进行研究，结合商务智能系统的层次结构，成都理工大学硕士学位论文 II 搭建了系统的层次结构，分为核心数据层、数据管理层、逻辑层及系统展示层； 分析了系统的功能流程， 。 （5）对 X 公司进行了实例分析。从企业信用管理组织结构，管理流程两方 面对企业实施赊销信用风险管理系统进行了分析。 利用 Structs+ Hibernate 技术实 现了 B/S
5、 架构的系统， 说明了本文提出的赊销信用风险管理系统的研究是有成效 的，它能够有效指导企业赊销信用管理的开展。 关键词：赊销信用管理 支持向量机 信用风险评估模型 管理系统Abstract III Research of Enterprise Credit Risk Management System based on support vector machine Introduction of the author: Liu Xiang, male, was born in December, 1983 whose tutor was Professor Liu Hongjun. He gr
6、aduated from Chengdu University of Technology in management science and engineering major and was granted the Master Degree in June, 2008. Abstract Credit is the basis of the market economy and the prerequisite of normal running of market economy。Credit risk is everywhere，so it is urgent for us to r
7、ecognize the essence and characteristics of credit and credit economy，improve our ability to recognize and avoid credit risk， enhance the level of enterprise credit management and adapt to the needs of credit economys development。 Credit sale risk is a kind of credit risk，it means that credit buying
8、 customer may not pay all the accounts which leads to the enterprise faces great financial risk. There are a lot of existed problems in enterprise credit sale risk management needs to be improved and resolved. This paper improves the existed research achievement ，establishes credit sale risk managem
9、ent frame according to the practice of present credit sale risk management and based on this frame，according to the process of credit sale risk management， realize that credit sale risk management model includes client screening model and dynamic supervision model of client credit risk .This paper a
10、lso build mathematic model using SVM method and a more effective credit risk evaluation model using advanced classified forecast method and analyze the application of the credit sale risk management system through the a company practice of credit management in machinebuild industry. The outcomes of
11、this paper are as follows： （1）We analyze the reasons of the current credit of Chinas enterprises and problems of credit risk management. Through that analysis, we have some enlightenments of business credit risk management. These are, the establishment of an independent credit management, business p
12、rocess reengineering, as well as the use of advanced information technology to establish credit risk management systems. （2）We introduces the current domestic and foreign general enterprise risk management framework, and establishes the enterprise credit risk management, according to the analysis of
13、 business credit management in the front of the thesis, Including the establishment of business credit management, the development of成都理工大学硕士学位论文 IV enterprise credit risk management policies, credit risk management process analysis and the structure of models based on the risk management processes.
14、 （3）We introduce the current method of credit evaluation, and establishes an model of credit risk evaluation on credit sale. We divide the indicators of credit risk evaluation into quantitative indicators and qualitative indicators. Quantitative indicators are the clients financial indicators, quali
15、tative indicators are mainly industry environment of the customers and internal management factors. We preprocess of the evaluated data of client. Quantitative data to take normalized, qualitative data on the use of fuzzy methods to quantify. （4）We research the corporate credit risk management syste
16、m and put up the hierarchical structure of the risk management system in combination with the hierarchical structure of business intelligence system. The structure is divided into the core data layer, data management layer, logic layer and system display layer We analyze the functional process of system . （5）From the respect of
