基于主成分和类聚算法的决策支持系统研究

Principal component and clustering algorithms based Decision Support System

紧密合当前助学贷款实施现状,将数据挖掘技术应用于国家助学贷款决策支持和国家助学贷款的风险管理。在充分调研分析基础上选取能衡量申请助学贷款学生贫困程度的指标体系,首选用主成分分析法对指标体系进行降维处理以便于助学贷款决策支持;其次采用近邻扩展聚类法构建贷款信用风险评级模型,用于贷款风险控制。并在此基础之上初步实现助学贷款决策支持系统设计和实现为相关部门提供决策支持。

According to the Practice of Student Loans,This paper applies data mining methods in the credit risk management and the Decision Support for National student loan. Based on lots of researches, the index system was selected to measuring poverty of students with financial difficulties. First, we adopt the principle components analysis to reduce the dimension of the index system, and then we use clustering algorithm to build a model for credit risk management. Based preparatory work this paper give preliminary design for DSS of National Student Loan.