DECISION THEORETIC PRINCIPLES AND POLICIES FOR NOTIFICATION

Patent №

US 6,513,026

Granted

2003-01-28

Filed 2000

Owner

MICROSOFT CORPORATION

AI components

3

nlp · kr · planning

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

09596348

Performance of a decision-theoretic analysis to determine which notifications as can be received from notification sources should be conveyed to the user, and via which modes of which notification sinks, is disclosed. A value can be determined for each mode of each notification sink, equal to an expected value of information contained within the notification, minus an expected cost of disruption to convey the notification via each mode of each sink, minus an expected value of the user independently learning the information contained with the notification without notification, and minus an actual cost of conveying the notification via each mode of each sink. If this value is greater than a predetermined conveyance threshold for any mode of any sink, then the notification is conveyed via the mode of the sink having the highest such value.

Natural languageKnowledge representationPlanningH04L 51/224G06Q 10/107H04L 51/214H04L 67/306H04L 67/51H04L 67/52H04L 67/54H04L 67/55+8 more

AI classification

Knowledge representation1.00
Planning1.00
Natural language0.99
AI hardware0.21
Machine learning0.18
Speech0.00
Vision0.00
Evolutionary computation0.00

Ownership

MICROSOFT CORPORATION

assignment · 112270791

Assignors

HORVITZ, ERIC J., KADIE, CARL M., HOVEL, DAVID O.

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC