首页> 外国专利> System method and computer program for credit evaluation using artificial neural network

System method and computer program for credit evaluation using artificial neural network

机译:人工神经网络信用评估系统方法及计算机程序

摘要

An input vector corresponding to the input data set by receiving an input data set including a value for at least one item of at least one item related to a financial service application and at least one item for an applicant using machine learning A hidden vector having a lower dimension than the input vector is generated from, and the hidden vector is corrected based on a hidden vector distribution, and a value for at least one item of the input data set is missing from the corrected hidden vector. A missing data predictor for outputting a reconstruction vector including a predicted value for the value; And a credit evaluation processing unit that performs credit evaluation on the application based on the reconstruction vector.
机译:通过接收输入数据集来与输入数据集相对应的输入向量,该输入数据集包括至少一项与金融服务应用程序相关的项中至少一项的值和至少一项针对使用机器学习的申请人的项的值。生成比输入矢量低的维,并且基于隐藏矢量分布对隐藏矢量进行校正,并且从校正后的隐藏矢量中缺少输入数据集的至少一项的值。丢失数据预测器,用于输出包括该值的预测值的重建矢量;以及信用评估处理单元,其基于重构向量对应用执行信用评估。

著录项

  • 公开/公告号KR102156757B1

    专利类型

  • 公开/公告日2020-09-16

    原文格式PDF

  • 申请/专利权人 DATA REPUBLIC INC.;

    申请/专利号KR20190119250

  • 发明设计人 엄성민;

    申请日2019-09-27

  • 分类号G06Q40/02;G06N20;

  • 国家 KR

  • 入库时间 2022-08-21 11:03:44

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号