Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
1999-06-15
2002-07-09
Trammell, James P. (Department: 2171)
Data processing: database and file management or data structures
Database design
Data structure types
C704S002000, C704S200000
Reexamination Certificate
active
06418440
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to interactive dialogue systems for human-machine communication, and more particularly to the automated generation of dialogues for such systems.
BACKGROUND OF THE INVENTION
Recently, there has been a growing interest in the general area of dialogue systems for human-machine communications. In such a system, the human, or “user,” interacts with the computer through one or more of the following mechanisms: typing responses and/or queries using a keyboard; speaking into an audio input device (in which case the system includes a “speech recognizer” for converting the user-supplied audio input into words, represented, for example, in the form of text); and using a mouse to select objects (such as, for example, words or phrases) on a computer screen. Any such system requires a module commonly referred to as a dialogue manager that defines a means for the computer system to recognize the user's request and then respond through some appropriate action, such as, for example, providing a particular audio speech output or updating a database in a particular way.
Most of the research and development in the field of dialogue systems has been in the area of “domain-specific” applications, wherein the system's dialogue manager has been specifically trained for a somewhat narrowly defined task, and has been previously set up by a skilled computer system design engineer. Examples have included, inter alia, weather reports, airline travel information systems, and banking services. In these cases, the user cannot define the application, being limited to the restricted options offered by the system.
More recently, however, a limited number of domain-independent systems have been described in which a user can cause the system to generate a dialogue for a particular (i.e., domain-specific) application automatically, without the need for a computer system design engineer. For example, U.S. Pat. No. 4,688,195, issued to Craig W. Thompson et al. (Hereinafter “Thompson et al.”), describes a “system for interactively generating a natural-language input interface, without any computer-skill programming work being required.” (See Thompson et al., abstract.) In particular, Thompson et al. describes, inter alia, a system which generates a user-customized natural language menu interface from a user-supplied database and from inputs supplied interactively by the user. More specifically, the system generates a domain-dependent context-free grammar by combining a domain-independent grammar and lexicon with a domain specification for a predetermined application.
However, in all of these prior art systems, including Thompson et al., the dialogue which is provided (whether “manually” generated by a skilled computer system design engineer, or “automatically” generated by a dialogue generation system based on a given user's input) is static. That is, once created, the dialogue is not responsive to the ever-changing external environment. For example, even when a dialogue is generated responsive to a user's desires and expectations, the available information upon which specifics of the dialogue may be based (e.g., the set of available documents retrieved in response to an appropriate search query posed over the Internet's World Wide Web) is fixed at the time the dialogue is generated. Thus, as the available external information changes, the dialogues generated by these prior art systems become rapidly outdated.
SUMMARY OF THE INVENTION
In accordance with the principles of the present invention, a customized method or algorithm for holding an interactive dialogue session between a (human) user and a machine (hereinafter referred to simply as a “dialogue”) is generated, such that the resulting dialogue advantageously responds to the user's requests and wherein the system's capability (i.e., the dialogue) is automatically modified thereafter based on dynamically changing external databases. Specifically, a computer system acts as a Dialogue Generator agent by creating such a customized dialogue consisting of services that are organized and presented in a form that is a combination of the user's expectations and the system's capabilities. In particular, the system's capabilities advantageously include the information content of database/service providers (such as, for example, a distributed information source such as the Internet's World Wide Web or a corporate file system), and the Dialogue Generator advantageously modifies the dialogue periodically in response to this dynamically changing external environment.
More specifically, the present invention consists of a method and apparatus for generating a human-machine dialogue comprising steps or modules for receiving a user request comprising at least one topic of interest; retrieving from an external database at a first point in time, a first set of documents comprising information relating to the at least one topic of interest; generating a human-machine dialogue based on the information comprised in said first set of documents; retrieving from the external database at a second point in time later than said first point in time, a second set of documents comprising information relating to the at least one topic of interest, the information comprised in said second set of documents having at least one difference from the information comprised in said first set of documents; and modifying the human-machine dialogue based on the information comprised in said second set of documents.
REFERENCES:
patent: 4688195 (1987-08-01), Thompson et al.
patent: 5694558 (1997-12-01), Sparks et al.
patent: 5748841 (1998-05-01), Morin et al.
patent: 5774860 (1998-06-01), Bayya et al.
patent: 5860059 (1999-01-01), Aust et al.
patent: 5886643 (1999-03-01), Diebboll et al.
patent: 5937422 (1999-08-01), Nelson et al.
patent: 6044347 (2000-03-01), Abella et al.
patent: 6065117 (2000-05-01), White
patent: 6081780 (2000-06-01), Lumelsky
patent: 6219643 (2001-04-01), Cohen et al.
patent: 6236968 (2001-05-01), Kanevsky et al.
patent: 6246981 (2001-06-01), Papineni et al.
patent: 6275225 (2001-08-01), Rangarajan et al.
patent: 0749 081 (1996-12-01), None
patent: 0848 338 (1998-06-01), None
“A Langugage For Creating Speech Applications” by Andrew Pargellis, Qiru Zhou, Antoine Saad, Chin-Hui Lee, 5thInternational Conference on Spoken Language Processing, Sydney Convention and Exhibition Centre 30 Nov.4,-Dec., 1998, pp. 1619-1622.
Thomas, CH G et al., “Using Agents to Personalize The Web”, IUI International Conf. On Intelligent User Interfaces, pp. 53-60 (1997).
Quintana, Y., “Knowledge-Based Information Filtering of Financial Information”, Proceedings of the National Online Meeting, pp. 279-285 (1997).
Ukelson, J. et al., “A dialogue Manager for Efficient adaptive Man-Machine Dialogues”, Proceedings of the Annual International Computer Software and Applications conference, pp. 588-595 (1989).
Kuo Hong-Kwang Jeff
Lee Chin-Hui
Pargellis Andrew Nason
Brown Kenneth M.
Lucent Technologies - Inc.
Trammell James P.
Wang Mary
LandOfFree
System and method for performing automated dynamic dialogue... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for performing automated dynamic dialogue..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for performing automated dynamic dialogue... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2881409