Multi-level decision-analytic approach to failure and repair...

Details Multi-level decision-analytic approach to failure and repair... Multi-level decision-analytic approach to failure and repair...

1999-06-04

2002-12-03

Baderman, Scott (Department: 2184)

Error detection/correction and fault detection/recovery

Data processing system error or fault handling

Reliability and availability

C704S251000, C704S257000

Reexamination Certificate

active

06490698

ABSTRACT:

FIELD OF THE INVENTION
This invention relates generally to computer-user interaction, and more particularly to a multi-level decision-analytic approach to failure and repair in such interaction.
BACKGROUND OF THE INVENTION
Generally, computer-user interaction has focused on the user conforming more to idiosyncrasies of the computer than vice-versa. For example, while users in noncomputer interactions (such as human-human interactions) typically communicate with a combination of verbal and nonverbal signals, this is generally not done with computer-user interactions. Rather, the user is forced to input information into a computer in a manner more easily understood by the computer—such as constrained voice inputs, text input from a keyboard, pointing, movement and clicking input from a mouse, etc. As a result, this imposed unnaturalness of the computer-user interface has played a part in hampering efforts to make computers easier to use and more an intuitive part of everyday life.
In human-human dialog, speakers and listeners elegantly coordinate the presentation and acceptance of utterances to achieve and confirm mutual understanding. In the process, they make decisions under uncertainty that minimize the risk of misunderstanding and contribute to furthering the goals of the activity. Uncertainty usually always plays a part in dialog. For example, a listener may be uncertain about the articulation of an utterance. Likewise, a speaker may be uncertain about the attentiveness or comprehension of the listener. Although participants may tolerate a small degree of uncertainty, an excessive amount in a given context can lead to misunderstanding along with all of its associated costs, such as the unwanted premature termination of a joint activity.
In human-computer dialog, the success of spoken interaction systems that integrate component technologies such as speech recognition, text to speech, and natural language generation, relies upon the precision of the component technologies. However, while uncertainty and imprecision characterize human-human dialog, people manage quite well in most situations. They display not only the ability to reason about key uncertainties and their costs, but they also exploit strategies such as grounding for collaboratively resolving them. Conversely, prior-art approaches to managing uncertainty within human-computer dialog usually perform much less well.
Prior art approaches to resolving uncertainty within computer-human dialog, such as uncertainty about the attentiveness and comprehension of the listener, are typically ad hoc. The prior art approaches generally do not reason about the sources of the uncertainty and how to resolve them. For example, they generally do not distinguish between communication failures due to poor fidelity of the communication channel or the user simply not being attentive at the time of the utterance execution. This makes for less robust and less flexible systems when uncertainty is encountered, and leads to a less natural contextual experience for the user.
For these and other reasons, there is a need for the present invention.
SUMMARY OF THE INVENTION
The invention relates to a multi-level decision-analytic approach to failure and repair within computer-user communications. In one embodiment, a computerized system repairs communication failure within a computer-user interaction context, and includes a maintenance module, an intention module, and a conversation control subsystem. The maintenance module manages uncertainty regarding signal identification and channel fidelity. The intention module is supported by the maintenance module, and manages uncertainty about the recognition of user's goals from signals. The conversation control subsystem surrounds both the modules, and manages the joint activity between the computer and the user, as well as one or more high-level events regarding the joint activity.
Thus, in one embodiment of the invention, each of the modules and the control subsystem manage uncertainty at different levels. The maintenance module manages uncertainty at the channel level and the signal level. The intention module manages uncertainty at the intention level. The conversational control subsystem manages uncertainty at the conversation level.
In this manner, embodiments of the invention provide for advantages over the prior art. By discerning where the uncertainty lies within a computer-user dialog, the inventive system is able to more naturally recover from any failure that may result from the uncertainty. For example, the manner by which repair is accomplished when uncertainty lies at the channel level—e.g., there is a failure between the basic link between the computer and the user—is different than when uncertainty lies at the intention level—e.g., the user is not understanding the computer's intentions, or viceversa. The end result is that the computer-user interaction experience is more natural for the user involved.
Embodiments of the invention include computer-implemented methods, computer-readable media, and computerized systems of varying embodiments. Still other embodiments, advantages and aspects of the invention will become apparent by reading the following detailed description, and by reference to the drawings.


REFERENCES:
patent: 5255386 (1993-10-01), Prager
patent: 5748841 (1998-05-01), Morin et al.
patent: 5864848 (1999-01-01), Horvitz et al.
patent: 6311159 (2001-10-01), Van Tichelen et al.
Araki M, Doshita S, Cooperative Spoken Language Model Using Bayesian Network and Event Hierarchy, IEICE Transaction on Information and Systems, vol. E78-D No. 6, June 1995, pp. 629-635, XP002148396, Japan.
Inspec Database, Institute of Electrical Engineers, Stevenage, GB, Inui K et al, A framework of decision-theoretic utterance planning, database accession No. 5757396, XP002148397, abstract, Journal of Japanese Society for Artificial Intelligence, Sep. 1997, Japanese Soc Artificial Intelligence, Japan, ISSN 0912-8085, vol. 12, pp. 760-769, 1997.
U.S. patent application Ser. No. 09/055,477, Platt, filed Apr. 6, 1998.
Eric Horvitz, David Heckerman, et al., Heuristic Abstraction in the Decision-Theoretic Pathfinder System, Proceedings of the Thirteenth Symposium on Computer Applications in Medical Care, IEEE Computer Society Press, 1989.
David Heckerman, Eric Horvitz, Inferring Informational Goals from Free-Text Queries: A Bayesian Approach, Fourteeneth Conference on Uncertainty in Artificial Intelligence, 1998.
Stephen D. Richardson, William B. Dolan, Lucy Vanderwende, MindNet: acquiring and structure semantic information from text, MSR-TR-98-23, Proceedings of the 17th International Conference on Computational Linguistics, May 29, 1998.
Stephen D. Richardson, Bootstrapping Statistical Processing into a Rule-based Natural Language Parser, MSR-TR-95-48, Proceedings of Workshop, The Balancing Act: Combining Symbolic and Statistical Approaches to Language, Jul. 1994.
Stephen D. Richardson, Lucy Vanderwende, William Dolan, Combining Dictionary-Based and Example-Based Methods for Natural Language Analysis, MSR-TR-93-08, Jun. 1993.
Goodwin, Between and within: Alternative sequential treatments of continuers and assessments, Human Studies, 1986(9), pp. 205-217.
Grice, Meaning, Philosophical Review, 1957 (66), pp. 377-388.
Grice, Logic and conversation, Syntax and Semantics 3: Speech Acts, 1975, pp. 41-58.
Jefferson, Side sequences, Studies in Social Interaction, 1972, pp. 294-338.
Schegloff and Sacks, Opening up closings, Semiotica, 1973, pp. 289-327.
Clark, Making sense of nonce sense, The Process of Language Understanding, 1983, pp. 297-331.
Cohen and Levesque, Preliminaries to a collaborative model of dialogue, Speech Communication, 1994 (15), pp. 265-274.
Clark and Wilkes-Gibbs, Referring as a Collaborative Process, chapter 23 of Intentions in Communication, 1990.
Platt, East Training of Support Vector Machines Using Sequential Minimal Optimization, chapter 12 of Advances in Kernel Methods: Support Vector Learning, 1999.
Clark and Brennan, Grounding in Communication, chapter 7 of Perspectives

Affiliated with

Also associated with

Rating

Multi-level decision-analytic approach to failure and repair... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Multi-level decision-analytic approach to failure and repair..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multi-level decision-analytic approach to failure and repair... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2917766