首页> 外国专利> Retirement risk learning model construction system, retirement risk learning model construction method and retirement risk learning model construction program

Retirement risk learning model construction system, retirement risk learning model construction method and retirement risk learning model construction program

机译:退休风险学习模型构建体系、退休风险学习模型构建方法和退休风险学习模型构建方案

摘要

To provide a resignation risk learning model construction system and a resignation risk learning model construction method which effectively assist personnel strategy of a company indirectly by providing a learning model to attain higher accuracy of prediction of a resignation risk of an employee.SOLUTION: A resignation risk learning model construction system 1 comprises: data group obtaining means 5 which obtains a data group regarding each existing employee or the like; incumbent and resignation data obtaining means which obtains an incumbent and resignation data group of each existing employee or the like; prediction model construction means group which construct a prediction model group for predicting a resignation risk when joining an employing company; and integrated prediction model construction means 7 which constructs an integrated prediction model for predicting a resignation risk of the existing employee or the like by substituting in an incumbent and resignation discriminant produced from combination of each integrated data group obtained by integrating the prediction model group and each incumbent and resignation label corresponding to the prediction model group.SELECTED DRAWING: Figure 1
机译:本发明提供一种离职风险学习模型构建系统和离职风险学习模型构建方法,通过提供学习模型来实现对员工离职风险的更高预测精度,从而间接有效地辅助公司的人事战略。解决方案:一种离职风险学习模型构建系统1包括:数据组获取装置5,其获取关于每个现有员工等的数据组;在职和离职数据获取装置,获取每个现有员工的在职和离职数据组等;预测模型构建是指构建预测模型组的组,用于预测加入雇佣公司时的离职风险;以及集成预测模型构建装置7,其通过替换由通过集成预测模型组和对应于以下各项的每个在职和离职标签而获得的每个集成数据组的组合产生的在职和离职判别式来构建用于预测现有员工等的离职风险的集成预测模型:预测模型组。所选图纸:图1

著录项

  • 公开/公告号JP7072755B2

    专利类型

  • 公开/公告日2022-05-23

    原文格式PDF

  • 申请/专利权人 株式会社アッテル;

    申请/专利号JP20190236340

  • 发明设计人 塚本 鋭;

    申请日2019-12-26

  • 分类号G06Q10/10;G06N20;G06Q10/06;

  • 国家 JP

  • 入库时间 2022-08-25 01:09:55

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