Method and system for generating optimized response to user input

Patent metadata

Field	Value
Publication number	US-10990602-B2
Application number	US-201816049886-A
Country	US
Kind code	B2
Filing date	Jul 31, 2018
Priority date	Jun 18, 2018
Publication date	Apr 27, 2021
Grant date	Apr 27, 2021

Abstract

The present disclosure relates to a method and a system for generating optimized response to user input. The system may receive a user input indicative of a data required by the user. The system identifies one or more keywords based on the user input. The system determines user expertise level based on search graphs generated using the one or more keywords. The system retrieves a plurality of responses relevant to the data based on the one or more keywords. The system assigns a value to each of the plurality of responses based on the user expertise level. The system identifies a base response in one or more responses having the value greater than a threshold value. Finally, the system collates content of the one or more responses excluding the base response with content of the base response, in a pre-defined sequential order, for generating the optimized response to user input.

First claim

What is claimed is:

1. A method for generating optimized response to user input, the method comprising:
receiving, by a response generation system, a user input indicative of data required by the user;
identifying, by the response generation system, one or more keywords based on the user input;
determining, by the response generation system, user expertise level based on a result of comparison of a search graph and a reference graph, wherein the search graph is generated based on the one or more keywords, and the reference graph is generated based on a domain associated with the data required by the user,
wherein the search graph indicates a link between the one or more keywords, the reference graph comprises potential keywords related to the domain associated with the data, represented in a sequential order and wherein the search graph and the reference graph are compared semantically for determining the user expertise level, and
wherein determining the user expertise level is further based on a plurality of user parameters, wherein a weighted sum of the plurality of user parameters and the result of the comparison of the search graph and the reference graph is used for determining the user expertise level;

retrieving, by the response generation system, a plurality of responses relevant to the data, from a database, based on the one or more keywords;
assigning, by the response generation system, a value to each of the plurality of responses based on the user expertise level, wherein the value associated with each response is indicative of a measure of relevancy of the corresponding response to the data required by the user;
identifying, by the response generation system, one or more responses having the value greater than a threshold value from the plurality of responses and characterizing one of the one or more responses as a base response; and
collating, by the response generation system, content of the one or more responses excluding the base response with content of the base response, in a pre-defined sequential order, for generating the optimized response to the user input.

2. The method as claimed in claim 1 , wherein the user expertise level is indicative of knowledge of the user in the domain associated with the data.

3. The method as claimed in claim 1 , wherein the one of the one or more responses is characterized as the base response when the value associated with the one of the one or more responses is greater than the value associated with the remaining of the one or more responses.

4. The method as claimed in claim 1 , wherein the one or more keywords comprises implicit keywords and explicit keywords, wherein the explicit keywords are directly derived from the user input and wherein the implicit keywords are derived based on pre-learnt domain knowledge and the content.

5. The method as claimed in claim 1 , wherein determining the user expertise level further comprises dynamically generating queries to the user based on the user input and responses received for the queries.

6. The method as claimed in claim 1 , wherein the plurality of user parameters comprises at least one of personal profile of the user, social media profile of the user, search history of the user, search patterns and choice of keywords used by the user.

7. A response generation system, for generating optimized response to user input, said response generation system comprising:
a processor; and
a memory, communicatively coupled with the processor, storing processor executable instructions, which, on execution causes the processor to:
receive, a user input indicative of data required by the user;
identify, one or more keywords based on the user input;
determine, user expertise level based on a result of comparison of a search graph and a reference graph, wherein the search graph is generated based on the one or more keywords, and the reference graph is generated based on a domain associated with the data required by the user,
wherein the search graph indicates a link between the one or more keywords, the reference graph comprises potential keywords related to the domain associated with the data, represented in a sequential order and wherein the search graph and the reference graph are compared semantically for determining the user expertise level, and
wherein determining the user expertise level is further based on a plurality of user parameters, wherein a weighted sum of the plurality of user parameters and the result of the comparison of the search graph and the reference graph is used for determining the user expertise level;
retrieve, a plurality of responses relevant to the data, from a database, based on the one or more keywords;
assign, a value to each of the plurality of responses based on the user expertise level, wherein the value associated with each response is indicative of a measure of relevancy of the corresponding response to the data required by the user;
identify, one or more responses having the value greater than a threshold value from the plurality of responses and characterizing one of the one or more responses as a base response; and
collate, content of the one or more responses excluding the base response with content of the base response, in a pre-defined sequential order, for generating the optimized response to the user input.

8. The response generation system as claimed in claim 7 , wherein the user expertise level is indicative of knowledge of the user in the domain associated with the data.

9. The response generation system as claimed in claim 7 , wherein the one of the one or more responses is characterized as the base response when the value associated with the one of the one or more responses is greater than the value associated with the remaining of the one or more responses.

10. The response generation system as claimed in claim 7 , wherein the one or more keywords comprises implicit keywords and explicit keywords, wherein the explicit keywords are directly derived from the user input and wherein the implicit keywords are derived based on pre-learnt domain knowledge and the content.

11. The response generation system as claimed in claim 7 , wherein determining the user expertise level further comprises dynamically generating queries to the user based on the user input and responses received for the queries.

12. The response generation system as claimed in claim 7 , wherein the plurality of user parameters comprises at least one of personal profile of the user, social media profile of the user, search history of the user, search patterns and choice of keywords used by the user.

13. A non-transitory computer readable medium including instructions stored thereon that when processed by at least one processor cause a response generation system to perform operations comprising:
receiving a user input indicative of data required by the user;
identifying one or more keywords based on the user input;
determining user expertise level based on a result of comparison of a search graph and a reference graph, wherein the search graph is generated based on the one or more keywords, and the reference graph is generated based on a domain associated with the data required by the user,
wherein the search graph indicates a link between the one or more keywords, the reference graph comprises potential keywords related to the domain associated with the data, represented in a sequential order and wherein the search graph and the reference graph are compared semantically for determining the user expertise level, and
wherein determining the user expertise level is further based on a plurality of user parameters, wherein a weighted sum of the plura

Assignees

• Wipro Ltd

Inventors

• Bangalore Narayanamurthy Vinutha
• Ramachandra Iyer Manjunath

Classifications

• G06F16/9024

  Graphs; Linked lists (G06F16/9027 takes precedence) · CPC title

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title

• G06F16/24578 Primary

  using ranking · CPC title

• G06F16/2453 Primary

  Query optimisation · CPC title