Alert differentiation device, alert differentiation method, and alert differentiation program

The alert detection device addresses the challenges of manual data entry and false positives by automatically generating inferred state information from notification data, improving alert differentiation accuracy and reducing the risk of information misuse.

JP7864020B2Active Publication Date: 2026-05-22MITSUBISHI ELECTRIC CORP
View PDF 4 Cites 0 Cited by

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
MITSUBISHI ELECTRIC CORP
Filing Date
2022-06-15
Publication Date
2026-05-22

AI Technical Summary

Technical Problem

Conventional alert detection systems require users to manually enter detailed behavioral changes into a database, which can lead to misuse of information by malicious insiders and are cumbersome to manage, and struggle with false positives due to flexible work arrangements and legitimate deviations from normal behavior.

Method used

An alert detection device that extracts status change information from notification data, acquires influence feature information from a database, generates inferred state information, and determines the likelihood of false positives based on this information to differentiate alerts without requiring manual entry of behavioral changes.

Benefits of technology

Enables accurate alert differentiation without manual data entry, reducing the risk of information misuse and improving efficiency by automatically generating detailed information for distinguishing alerts, and enhancing accuracy through pre-configured databases and feedback mechanisms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007864020000001
    Figure 0007864020000001
  • Figure 0007864020000002
    Figure 0007864020000002
  • Figure 0007864020000003
    Figure 0007864020000003
Patent Text Reader

Abstract

To enable a user to distinguish an alert even without describing details of changes of behaviors accompanying a situation change in a database.SOLUTION: A situation change detection unit 111 extracts situation change information from notification data notifying a situation change of a user. An influence feature acquisition unit 112 acquires influence feature information of a situation keyword corresponding to the situation change shown in the situation change information from an influence feature database. A predicted state generation unit 113 generates predicted state information about the user by using the situation change information and the acquired influence feature information. An alert distinguishing unit 122 determines erroneous detection probability of the generated alert on the basis of the predicted state information when an alert to a behavior of the user is generated.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art