Response generator for mimicking human-computer natural language conversation
First Claim
Patent Images
1. An autonomous response method, comprising:
- automatically updating a context database, said context database containing one or more context elements selected from the group consisting of a score in a sporting contest, a value of a market index, a value of a commodity, a result of a poll and a result of a survey, or some combination thereof;
using one or more of said context elements to generate a learned mood number (Ml);
automatically updating a statement-response database, said updating including associating and storing said learned mood number with said response;
receiving a natural language query;
automatically generating at least two possible responses to said natural language query;
automatically obtaining current values of said context elements corresponding to those used in generating said learned mood number;
using said current values of said context elements to generate a current mood number (Mc);
automatically weighting said possible responses using said learned mood value stored with said response and said current mood value using the formula;
weight=1/(1+C|(Mc)−
(Ml)|), where C is a constant related to a suitability of said possible response and | | indicates the absolute difference between the current and learned mood values; and
,automatically using a processor selecting said lowest weighted response to generate a natural language response to said natural language query.
0 Assignments
0 Petitions
Accused Products
Abstract
The present invention is an autonomous response engine and method that can more successfully mimic a human conversational exchange. In an exemplary, preferred embodiment of the invention, the response engine has a statement-response database that is autonomously updated, thus enabling a database of significant size to be easily created and maintained with current information. The response engine autonomously generates natural language responses to natural language queries by following one of several conversation strategies, by choosing at least one context element from a context database and by searching the updated statement-response data base for appropriate matches to the queries.
449 Citations
16 Claims
-
1. An autonomous response method, comprising:
-
automatically updating a context database, said context database containing one or more context elements selected from the group consisting of a score in a sporting contest, a value of a market index, a value of a commodity, a result of a poll and a result of a survey, or some combination thereof; using one or more of said context elements to generate a learned mood number (Ml); automatically updating a statement-response database, said updating including associating and storing said learned mood number with said response; receiving a natural language query; automatically generating at least two possible responses to said natural language query; automatically obtaining current values of said context elements corresponding to those used in generating said learned mood number; using said current values of said context elements to generate a current mood number (Mc); automatically weighting said possible responses using said learned mood value stored with said response and said current mood value using the formula; weight=1/(1+C|(Mc)−
(Ml)|), where C is a constant related to a suitability of said possible response and | | indicates the absolute difference between the current and learned mood values; and
,automatically using a processor selecting said lowest weighted response to generate a natural language response to said natural language query. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. An autonomous response apparatus, comprising:
- a processor implementing the steps of;
automatically updating a context database, said context database containing one or more context elements selected from the group consisting of a score in a sporting contest, a value of a market index, a value of a commodity, a result of a poll and a result of a survey, or some combination thereof; using one or more of said context elements to generate a learned mood number (Ml); automatically updating a statement-response database, said updating including associating and storing said learned mood number with said response; receiving a natural language query; automatically generating at least two possible responses to said natural language query; automatically obtaining current values of said context elements corresponding to those used in generating said learned mood number; using said current values of said context elements to generate a current mood number (Mc); automatically weighting said possible responses using said learned mood value stored with said response and said current mood value using the formula; weight=1/(1+C|(Mc)−
(Ml)|), where C is a constant related to a suitability of said possible response and | | indicates the absolute difference between the current and learned mood values; and
,automatically selecting said lowest weighted response to generate a natural language response to said natural language query. - View Dependent Claims (11, 12, 13, 14, 15, 16)
- a processor implementing the steps of;
Specification