Data processing: artificial intelligence – Neural network – Learning task
Reexamination Certificate
1998-05-13
2001-07-17
Davis, George B. (Department: 2122)
Data processing: artificial intelligence
Neural network
Learning task
C706S002000, C706S050000
Reexamination Certificate
active
06263326
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to computerized state machines, and more specifically may apply to state machines having probabilistic transitions among states, where the machine is subjected to external stimulus.
BACKGROUND
Many expert systems have been developed to model human analytical reasoning. Also, many machine learning techniques exist for modeling human learning ability. Such systems and techniques have not generally incorporated the modeling of emotion, and uncertainty which may accompany transitions from one emotional state to another. Further, such systems and techniques have not generally included an aspect of emotional states being modulated under conditions of uncertainty responsive to emotion bearing events. A need therefore exists in this area.
More generally, a need exists for determining a state of a generalized system under conditions of uncertainty, particularly where the system is subject to an external stimulus which modifies uncertainty parameters.
SUMMARY
The foregoing need is addressed in the present invention.
It is one object of the invention to model system states under conditions of uncertainty. The modeling uses a computational engine. In one implementation, the modeling includes some aspects of Markov modeling.
A primary application contemplated for such a model is in the area of human emotion. In such an application, the system states represent feelings and states of mind or body (referred to herein collectively as emotional states), and the uncertainty concerns uncertainty associated with changes in emotional state. Accordingly, it is another object to include emotional components in the intelligence of “intelligent agents”, such as may be used in a variety of applications.
This system and method can be applied as a solution in various service industries. Some examples are:
1. Flight Controller Room Monitoring (Aviation)
Referring to
FIG. 7
, an emotion engine
710
monitors the mental and emotional states of a flight control crew based on stimuli collected about the traffic, close calls, time of day, length on desk, etc., which is input to the engine
710
from stations
720
,
730
, etc. Output from the emotion engine
710
triggers indicators
750
,
760
, etc. at the flight controller stations and provides mental and emotional state status to the shift schedule console
780
. Additionally, a similar emotion engine and data collection system could also be used to monitor a pilot's emotional state.
2. Medical Patient Monitoring (Medical)
In this application, the system monitors the emotional state of medical patients. Certain information, such as heart rate, blood pressure, etc. is automatically collected and input to the system. Other information, such as delivery of medication, physical examination observations, etc. are input manually to the system. The system gives a health care provider an automated way to monitor emotional state, chart progress, and recommend treatment.
3. Personal Calendar Assistant (Planning)
Referring to
FIG. 8
, emotional engine
810
assists in scheduling meetings to optimize on the probability of a person being in a receptive state. The emotion engine
810
receives, as input, data from previous calendar entries including the current day and Day−1, Day−2, and Day−3. The engine
810
outputs a predicted emotional state for proposed future meetings on the current day or future days, such as Day+1, etc. This helps manage calendars, including helping administrative assistants schedule or reject meetings.
4. Media Application (Advertising)
In this application, the system is used to judge the effectiveness of a movie, TV show, or commercial by measuring the change in emotional state evoked due to the media event.
5. Cell Phone Monitoring (Communication)
In appropriate business applications, when caller privacy concerns are not preemptive, a cell phone center or dispatch center monitors call content, call frequency, caller tone of voice, receiver tone of voice, etc. to determine a “nervousness index”, alertness index, etc. For example, police, delivery, and taxicab dispatch applications are especially suitable.
According to the present invention, the foregoing and other objects are attained by a computer aided method for modulating among a number of predetermined states. Information is stored in a memory associated with the computer and processed by the computer. The stored information includes information representing predefined categories of an external stimulus, n predetermined states, and likelihood functions for probability of respective transitions among the respective states. Transition probability values are computed, using the likelihood functions, in response to categorization of the external stimulus which is input to the computer. In a further aspect, the system transitions from an initially activated one of the emotional states to subsequently activated ones of the states, in response to the transition probability values.
The invention further contemplates that the states may represent emotional states, such as emotional states of a human or other sentient being or group of beings, or of an entity to which an emotional state may be imputed.
In addition, in one embodiment the external stimulus includes information representing emotion bearing events, and initializing the system includes defining: i) potential ones of such events and ii) one or more effects, on the transition probabilities, of the occurrence of such a defined potential event. That is, for each of the emotion bearing events first, second, third, etc. emotional characteristics are defined, and an effect on parameters of the likelihood functions is defined for each of the characteristics. According to this embodiment, the computing of probabilities responsive to the external stimulus may include computing the probabilities responsive to: i) an actual occurrence of one of the defined potential events, and ii) the defined effects.
In alternative embodiments, the invention may be practiced as a computer system, or as a computer program product for performing the above described method steps.
It is an advantage of the present invention that it can be applied as a solution in various service industries like the health, aviation, advertising, and communication industries.
Additional objects, advantages, and novel features are set forth in the following description, or will be apparent to those skilled in the art or those practicing the invention. Other embodiments are within the spirit and scope of the invention. These objects and embodiments may be achieved by the combinations pointed out in the appended claims. The invention is intended to be limited only as defined in the claims.
REFERENCES:
patent: 5289562 (1994-02-01), Mizuto et al.
patent: 5381513 (1995-01-01), Tsuboka
patent: 5459798 (1995-10-01), Bailey et al.
patent: 5774591 (1998-06-01), Black et al.
patent: 5884626 (1999-03-01), Kuroda et al.
patent: 5911581 (1999-06-01), Reynolds et al.
patent: 5918222 (1999-06-01), Fukui et al.
patent: 5987415 (1999-11-01), Breese et al.
patent: 6020883 (2000-02-01), Herz et al.
Chandra et al, “Using Causal Reasoning to Validate Stochastic Models” IEEE Proceedings of the 10th Conference on Artificial Intelligence for Applications, Mar. 1994.*
Yang et al, “Individual Cell Loss Probabilities and Background Effects in ATM Networks”, IEEE International Conference on Communications, May 1993.*
William J. Stewart; 1994; Introduction to the Numerical Solution of Markov Chains; “Direct Methods”, Princeton University Press; pp. xv-xvii; 63-70.
Kishor Shridharbhai Trivedi; 1982; Probability and Statistics with Reliabiltiy, Queuing, and Computer Science Applications; “Discrete-Parameter Markov Chains”; Printice-Hall, Inc; Chapter 7, pp. 309-317.
Carver A. Mead; 1980; Introduction to VLSI Systems; “Algorithms for VLSI Processor Arrays”, Addison-Wesley; Chapter 8, pp. 271-292.
Arun Chandra, “A Computational Architecture to Model Human Emotions”, IEEE Proceedings of the Intelligent Information System, Dec. 1997.
Carwell Robert M.
Davis George B.
England Anthony V. S.
International Business Machines - Corporation
LandOfFree
Method product ‘apparatus for modulations’ does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method product ‘apparatus for modulations’, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method product ‘apparatus for modulations’ will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2451506