贝叶斯分类器在农村金融数据分类中的应用研究  

Application of Bias Classifier in Rural Finance in Data Classification

作　　者：蒲京京 苏巧荣 张朝瑜 肖振东 吴养会[1] Pu Jingjing(College of Science,Northwest A＆F University,Yangling 712100,China)

机构地区：[1]西北农林科技大学理学院,陕西杨凌712100 [2]西北农林科技大学经济管理学院,陕西杨凌712100

出　　处：《安徽农学通报》2018年第11期2-4,20,共4页Anhui Agricultural Science Bulletin

基　　金：2016年西北农林科技大学大学生创新训练项目(编号:201610712089)

摘　　要：针对农户产权抵押贷款信用问题,利用陕西高陵、临潼、杨陵3个地区农村金融抵押贷款的有关调研数据,运用贝叶斯网络分类模型进行分类研究,以农户所在地区、家庭类型、教育程度等10个因素作为自变量,以产权抵押贷款的意愿作为分类依据,结果表明,采用十折交叉验证的贝叶斯网络分类器对解决此类问题具有良好的分类效果,其分类结果可以为农户抵押贷款研究提供理论参考。According to the problem of farmers＇ property mortgage credit,the Bayes Network classification model was constructed which based on the investigation data related to the property mortgage in Gaoling,Lingtong and Yangling.We chosed the region,family type,level of education and other 7 factors as the independent variables,the property mortgage loan willingness as the independent variables.The analysis results show that Bayes Network classificationmodel which used ten-fold cross validation has good effect in solving these problems.Besides,the classification results provided the certain reference role in the household mortgage policy.

关 键 词：农村金融 贝叶斯网络分类器 ROC曲线 交叉验证 

分 类 号：F832[经济管理—金融学]