首页> 外国专利> INTEGRATED MONITORING SYSTEM FOR PERSONAL INFORMATION SECURITY PRODUCT

INTEGRATED MONITORING SYSTEM FOR PERSONAL INFORMATION SECURITY PRODUCT

机译:个人信息安全产品集成监控系统

摘要

The present invention relates to an integrated control system for personal information security products, comprising a business-based data conversion unit that converts data using personal information as business actions of a user into business action list data, and a business action list obtained from the data conversion unit for each user. Task-based clustering to form a task-based cluster by clustering the task-based data extraction section to be extracted, the task-based vector conversion section to vectorize the task action list extracted for each user, and the vectorized task action list using the K-average algorithm. A processing unit, a task-based threshold calculation unit that calculates a distance value threshold between the center of the task-based cluster and the elements using a t-digest algorithm, and a clustering that stores the task-based cluster and the distance threshold in a column-type data format. A profiling cluster storage module composed of a storage unit; It is characterized by consisting of an abnormal pattern analysis module including an abnormal pattern determination unit that determines whether or not abnormal behavior by comparing the analysis distance value of the cluster obtained from the user's subsequent personal information access behavior and the task-based distance value threshold obtained from the storage module. It is possible to reduce the rate of positive or negative errors by analyzing the user's work behavior on the generated and collected logs without fixing the abnormal pattern in advance.
机译:本发明涉及一种用于个人信息安全产品的集成控制系统,包括基于业务的数据转换单元,其使用个人信息作为用户的业务动作转换为业务动作列表数据的业务动作,以及从数据获得的业务动作列表每个用户的转换单元。基于任务的群集通过群集要提取的基于任务的数据提取部分来形成基于任务的群集,基于任务的向量转换部分向矢量化为每个用户提取的任务操作列表,以及使用的矢量化任务动作列表K平均算法。处理单元,基于任务的阈值计算单元,其使用T-Digest算法计算基于任务基群集群的中心和元素之间的距离值阈值,以及存储基于任务的群集的群集和距离阈值以列型数据格式。分析集群存储模块由存储单元组成;其特征在于由异常模式分析模块组成,包括异常模式确定单元,该异常模式确定单元通过比较从用户随后的个人信息访问行为和基于任务的距离值阈值获得的群集的分析距离值来确定异常行为从存储模块获得。通过分析用户的工作行为在未提前修复异常模式,可以通过分析用户的工作行为来降低正面或负误差的速率。

著录项

  • 公开/公告号KR20210028952A

    专利类型

  • 公开/公告日2021-03-15

    原文格式PDF

  • 申请/专利权人 주식회사 에스링크;

    申请/专利号KR1020190110142

  • 发明设计人 공병철;

    申请日2019-09-05

  • 分类号G06F21/55;G06F21/57;

  • 国家 KR

  • 入库时间 2022-08-24 17:42:25

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