Decision theoretic principles and policies for notification

Details Decision theoretic principles and policies for notification Decision theoretic principles and policies for notification

2000-06-17

2003-01-28

Starks, Jr., Wilbert L. (Department: 2121)

Data processing: artificial intelligence

Knowledge processing system

Knowledge representation and reasoning technique

C707S793000, C707S793000

Reexamination Certificate

active

06513026

ABSTRACT:

FIELD OF THE INVENTION
This invention relates generally to unified receipt and notification of alerts generated by varied devices and applications for conveyance to a user, and more particularly to performing decision-theoretic notification for utilization with such unified alert receipt and notification.
BACKGROUND OF THE INVENTION
Many computer users today receive information from a number of different sources, and utilize a number of different devices in order to access this information. For example, a user may receive e-mail and instant messages over a computer, pages over a pager, voice-mail over a phone, such as a cellular (“cell”) or landline phone, news information over the computer, etc. This makes it difficult for the user to receive all his or her different information, referred to also as alerts or notifications, wherever the user happens to be.
For example, a user may be away from his or her computer, but receive an important e-mail. The user may have access only to a cell phone or a pager, however. As another example, the user may be working on the computer, and have turned off the ringer and voice-mail indicator on the phone. When an important voice-mail is left, the user has no way of receiving this information on the computer.
Moreover, many of the alerts may not be important to the user—for example, an e-mail from the user's manager or co-worker should often receive higher priority than the latest sports scores. More generally, the value of the information contained in an alert should be balanced with the costs associated with the disruption of the user by an alert. Both the costs and value may be context sensitive. Beyond notifications about communications, users are alerted with increasing numbers of services, error messages, and computerized offers for assistance, generated by systems and applications running on the local computer or on external computer and communication systems.
SUMMARY OF THE INVENTION
This invention relates to performing 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. In one embodiment, a value is 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. The invention includes computer-implemented methods, machine-readable media, computerized systems, and computers of varying scopes. Other aspects, embodiments and advantages of the invention, beyond those described here, will become apparent by reading the detailed description and with reference to the drawings.


REFERENCES:
patent: 5864848 (1999-01-01), Horvitz
patent: 6021403 (2000-01-01), Horvitz
patent: 6421655 (2002-07-01), Horvitz et al.
U.S. application Ser. No. 09/007,894, Horvitz, A Technique for Prefetching, filed Jan. 15, 1998.
U.S. application Ser. No. 09/055,477, Methods and Apparatus for . . . , filed Apr. 6, 1998.
U.S. application Ser. No. 09/364,522, Horvitz, Methods for Display . . . , filed Jul. 30, 1999.
U.S. application Ser. No. 09/364,528, Horvitz, Methods for Routing . . . , filed Jul. 30, 1999.
U.S. application Ser. No. 09/365,293, Horvitz, Integration . . . , filed Jul. 30, 1999.
U.S. application Ser. No. 09/365,287, Horvita, A Computational . . . , filed Jul. 30, 1999.
U.S. application Ser. No. 09/364,527, Horvitz, Method for Automatically . . . , filed Jul. 30, 1999.
M. Sahami et al, (1998), A Bayesian approach to filtering junk email, in Workshop on learning for text categorizations, AAAI Technical Report WS-98-05, AAAI.
D. Koller et al, (1996) Toward optimal feature selection, in proceedings of 13thconference on machine learning, pp. 284-292, Morgan Kaufmann, San Francisco.
E. Horvitz et al, (1998) The Lumiere project, Bayesian user modeling for inferring the goals and needs of software users, in proceedings of the 14thconf on uncertainty in Al, pp. 256-265, Morgan Kaufmann, San Francisco.
J. Platt (1999), Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods, in Advances in Large Margin Classifiers, MIT Press, Cambridge, MA.
H. Leiberman (1995), Letizia, An agent that assists web browsing, in proceedings of IJCAI-95, Montreal Canada, Morgan Kaufmann, San Francisco.
Horvitz et al, (1995) Display of information for time-critical decision making, in proceedings of the 11thconf on uncertainty in Al, pp. 296-305, Monetrea, Canada.
M. Czerwinski et al (1999), Visualizing implicit queries for information management and retrieval, in proceedings of CHI'99, ACM SIGCHI Conf on informational and knowledge management, pp. 560-567, ACM.
S. Dumais et al (1998), Inductive learning algorithms and representations for text categorization, in Proceedings of 7thIntl Conf on Information and Knowledge Management, pp. 148-155, ACM.
Platt (1999), Fast training of support vector machines using sequential minimal optimization, in advances in kernel methods, support vector learning, MIT Press, Cambridge, MA.
Horvitz (1999), Principles of mixed-initiative user interfaces, in Proceedings of CHI'99, ACM SIGCHI Conf on Human Factors in Computing Systems, Pittsburgh, PA, 159-166, ACM.
Breese et al (1998) Empirical analysis of predictive algorithms for collaborative filtering, in Proceedings of the 14thconf on uncertainty in Al, pp. 43-52, AUAI, Morgan Kaufmann, San Francisco.
Horvitz, Rutledge (1991), Time dependent utility and action under uncertainty, in proceedings of 7thconf on uncertainty in Al, LA, CA, pp. 151-158, Morgan Kaufmann, San Francisco.
Horvitz, Seiver (1977), Time-critical action: representations and application, in proceedings of the 13thconf on uncertainty in Al (UAI-97), pp. 250-257, Providence, RI, Morgan Kaufmann, San Francisco.

Affiliated with

Also associated with

Rating

Decision theoretic principles and policies for notification does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Decision theoretic principles and policies for notification, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Decision theoretic principles and policies for notification will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3041983