US 11,238,100 B2Granted

Adapting conversational agent communications to different stylistic models

AssigneeInternational Business Machines Corporation

InventorsMarisa Affonso Vasconcelos, Paulo Rodrigo Cavalin, Claudio Pinhanez, Heloisa Caroline De Souza Pereira Candello

Life of this patent

Estimating the term…

patentgrep

Loading patent…

Abstract

A computer system includes a processor configured to execute a conversational agent generator function. The conversational agent generator function is configured to receive a plurality of model input queries. Each model input query includes an intent and a corresponding answer to the intent. A database stores a corpus including sets of textual responses indicative of a common meaning. Each textual response in a given set of textual responses indicates a common meaning constructed in a different style indicative of a corresponding tone of communication. The processor is further configured to execute a rephrasing function configured determine the answer of one of the plurality of model input queries, and convert the answer of one of the plurality of model input queries into a plurality of different rephrased answers from the corpus stored in the database. The processor further outputs the plurality of different rephrased answers to the conversational agent generator function.

Claims18

Dependent claims

A computer system comprising:

a processor in electrical communication with a database;
the processor configured to execute a conversational agent generator function,
wherein the conversational agent generator function is configured to receive a plurality of model input queries, each model input query including an intent and a corresponding answer to the intent,
wherein the database stores a corpus including sets of textual responses indicative of a common meaning, each textual response in a given set of textual responses indicative of a common meaning constructed in a different style indicative of a corresponding tone of communication, and
wherein the processor is further configured to execute a rephrasing engine configured to:
determine the answer of one of the plurality of model input queries;
convert the answer of one of the plurality of model input queries into a plurality of different rephrased answers from the corpus stored in the database;
and
output the plurality of different rephrased answers to the conversational agent generator function,
wherein the conversational agent generator computing system is further configured to perform operations comprising:
receive at least one request to update a current tone of communication of a local conversational agent included in at least one user device located remotely from the conversational agent generator computing system;
generate at least one updated model input query including a rephrased answer corresponding to an initial answer included in an initial model input query, the rephrased answer having a different tone of communication with respect to an initial tone of communication of the initial answer;
and
output the updated model input query including a rephrased answer to the user device,
wherein the at least one user device updates the local conversational agent based on the updated model input query including a rephrased answer.

12 more elements

A computer-implemented method of adapting answers generated by a conversational agent to different stylistic models, the method comprising:

receiving, at a conversational agent generator computing system, a plurality of model input queries, each model input query including an intent and a corresponding answer to the intent;
storing, via a knowledge database, a corpus including sets of textual responses indicative of a common meaning, each textual response in a given set constructed in a different style indicative of a corresponding tone of communication;
and
determining, via a rephrasing engine in electrical communication with the conversational agent generator and the knowledge database, the answer of a given model input query;
converting, via the rephrasing engine, the answer into a plurality of different rephrased answers from the corpus stored in the knowledge database;
and
outputting, via the rephrasing engine, the plurality of different rephrased answers to the conversational agent generator;
receiving, via the conversational agent generator computing system, at least one request to update a current tone of communication of a local conversational agent included in at least one user device located remotely from the conversational agent generator computing system;
generating, via the conversational agent generator computing system, at least one updated model input query including a rephrased answer corresponding to an initial answer included in an initial model input query, the rephrased answer having a different tone of communication with respect to an initial tone of communication of the initial answer;
outputting, via the conversational agent generator computing system, the updated model input query including a rephrased answer to the user device;
and
updating, via the at least one user device, the local conversational agent based on the updated model input query including a rephrased answer.

9 more elements

A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform operations comprising:

receiving, at a conversational agent generator computing system, a plurality of model input queries, each model input query including an intent and a corresponding answer to the intent;
storing, via a knowledge database, a corpus including sets of textual responses indicative of a common meaning, each textual response in a given set constructed in a different style indicative of a corresponding tone of communication;
determining, via a rephrasing engine in electrical communication with the conversational agent generator and the knowledge database, the answer of a given model input query;
converting, via the rephrasing engine, the answer into a plurality of different rephrased answers from the corpus stored in the knowledge database;
and
outputting, via the rephrasing engine, the plurality of different rephrased answers to the conversational agent generator,
receiving, via the conversational agent generator computing system, at least one request to update a current tone of communication of a local conversational agent included in at least one user device located remotely from the conversational agent generator computing system;
generating, via the conversational agent generator computing system, at least one updated model input query including a rephrased answer corresponding to an initial answer included in an initial model input query, the rephrased answer having a different tone of communication with respect to an initial tone of communication of the initial answer;
outputting, via the conversational agent generator computing system, the updated model input query including a rephrased answer to the user device;
and
updating, via the at least one user device, the local conversational agent based on the updated model input query including a rephrased answer.

8 more elements

Dependencies are inferred from the claim text.