基于神经网络评估模型的掌上贷款系统设计与实现

发布时间：2018-03-18 05:12

本文选题：小微贷款　切入点：贷款风险评价　出处：《杭州电子科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：中小企业对我国的经济发展具有重要作用，然而由于我国现阶段金融市场以及社会信用体系建设还不完善加之中小企业经营难以监管的特点，导致中小企业普遍存在融资困难的局面，严重限制了中小企业的发展。对于以上问题，本文首先对中小企业的经营特点以及银行贷款业务的流程进行了分析，找出影响银行对中小企业贷款积极性的主要原因，针对这些影响因素，在基于银行现有贷款业务流程的基础上，通过简化小微贷款业务流程并将其移植到移动客户端，来减小银行在小微贷款中的成本投入和提高客户经理的工作效率。同时针对中小企业贷款风险难控的问题，本文首先对现有预测模型建模方法进行了研究，根据不同建模方法的使用环境，结合本文评价模型的输入是以定量数据为主，，加之少量定性数据的特点,选择采用BP神经网络来构建风险评价模型为银行贷款提供决策支持。然后在分析了小微贷款企业的经营特点、影响小微贷款企业还款能力主要因素以及某银行贷款评价指标的基础上，以工业企业为例制定了本文评价模型的评价指标体系，同时针对BP网络具有的局部极小化问题，收敛速度慢等问题，文中又对BP网络算法做了部分改进，最后通过Matlab对评价模型进行了训练及测试，证明该模型具有一定的实用性。
[Abstract]:Small and medium-sized enterprises play an important role in the economic development of our country. However, due to the imperfect construction of financial market and social credit system in our country at present, it is difficult to supervise the management of small and medium-sized enterprises. As a result, SMEs generally have financing difficulties, which seriously limit the development of SMEs. For the above problems, this paper first analyzes the management characteristics of small and medium-sized enterprises and the process of bank loan business, finds out the main reasons that affect the enthusiasm of banks to SMEs loans, and aims at these factors. Based on the existing loan business process of the bank, we can reduce the cost investment and improve the work efficiency of the account manager by simplifying the micro-loan business process and transplanting it to the mobile client. At the same time, aiming at the problem of difficult to control the loan risk of small and medium-sized enterprises, this paper first studies the existing forecasting model modeling methods, according to the use environment of different modeling methods, combined with the evaluation model of this paper, the input of the model is based on quantitative data. Combined with the characteristics of a small amount of qualitative data, BP neural network is used to build a risk assessment model to provide decision support for bank loans. Then, the management characteristics of small and micro loan enterprises are analyzed. On the basis of the main factors affecting the repayment ability of small and micro loan enterprises and the evaluation index of a certain bank loan, the evaluation index system of the evaluation model is established for industrial enterprises as an example. At the same time, the local minimization problem of BP network is pointed out. In this paper, the BP neural network algorithm is partly improved. Finally, the evaluation model is trained and tested by Matlab, and the model is proved to be practical.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F276.3;F832.4;TP183

【参考文献】