Systems and methods for identifying a meaning of an ambiguous term in a natural language query

• US 10,169,470 B2
• Filed: 04/11/2016
• Issued: 01/01/2019
• Est. Priority Date: 04/11/2016
• Status: Active Grant

• Current Assignee: Rovi Guides, Inc. (Adeia Inc.)
• Original Assignee: Rovi Guides, Inc. (Adeia Inc.)
• Inventors: Malhotra, Manik, Mohaideen P, Ahmed Nizam, Venkataraman, Sashikumar
• Primary Examiner(s): Le, Uyen T
• Application Number:
  US15/095,833
• Publication Number:
  US 20170293690A1
• Time in Patent Office: 995 Days
• Field of Search: None
• CPC Class Codes: G06F 16/2455 Query execution
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

Methods and systems for identifying a meaning of an ambiguous term in a natural language query. The media guidance application isolates first and second terms from a query received from a user and identifies, in a knowledge graph, first and second pluralities of candidate components associated with the first and second terms. The first and second terms each having multiple candidate components indicates the first and second terms have ambiguous meanings. The media guidance application matches each candidate component of the first and second pluralities of candidate components to form a plurality of pairs and determines strength of association for each pair in the plurality of pairs. The media guidance application filters the plurality of pairs by strength of association for each pair and determines a plurality of possible meanings based on the filtered plurality of pairs. The media guidance application selects a meaning from the plurality of possible meanings.

Claims

1. A method for identifying a meaning of an ambiguous term in a natural language query, the method comprising:

receiving, via a user interface, a query from a user;

isolating a first term and a second term from the query received from the user;

accessing, at a storage device, a knowledge graph;

identifying, in the knowledge graph, a first plurality of candidate components associated with the first term, wherein the first term having more than one candidate component indicates the first term has an ambiguous meaning;

identifying, in the knowledge graph, a second plurality of identified candidate components associated with the second term, wherein the second term having more than one candidate component indicates the second term has an ambiguous meaning;

matching each candidate component of the first plurality of candidate components to each candidate component of the second plurality of candidate components to form a plurality of pairs;

determining a strength of association for each pair in the plurality of pairs based on a distance on the knowledge graph between two nodes of the respective pair;

filtering the plurality of pairs by the strength of association for each pair;

determining a plurality of possible meanings associated with the query based on the filtered plurality of pairs; and

selecting a meaning from the plurality of possible meanings associated with the query based at least in part on a ranking of the plurality of possible meanings.

2. The method of claim 1, further comprising:

isolating a command term from the query received from the user;

determining the plurality of possible meanings associated with the query based on the filtered plurality of pairs and the command term.

3. The method of claim 2, wherein determining the plurality of possible meanings further comprises:

determining a plurality of possible meanings associated with the command term, wherein each of the possible meanings is associated with expected terms;

comparing the expected terms associated with each possible meaning associated with the command term to the filtered plurality of pairs;

filtering out a subset of the plurality of pairs based on the comparison; and

determining the plurality of possible meanings associated with the query based on the subset of the plurality of pairs and the expected terms.

4. The method of claim 1, wherein filtering the plurality of pairs further comprises:

identifying a first pair of the plurality of pairs;

determining whether the first pair has a strength of association which is below a required threshold;

in response to determining that the strength of association is below the required threshold, excluding the first pair from the filtered plurality of pairs; and

in response to determining that the strength of association is not below the required threshold, including the first pair in the filtered the plurality of pairs.

5. The method of claim 4, further comprising:

determining the required threshold based on the strength of association for each of the plurality of pairs;

adjusting the required threshold based on the filtered plurality of pairs; and

filtering the plurality of pairs based on the adjusted required threshold.

6. The method of claim 4, further comprising:

receiving feedback on the accuracy for the plurality of possible meanings associated with the query; and

adjusting the required threshold based on the received feedback.

7. The method of claim 1, further comprising:

isolating a third term from the query received from the user;

determining the plurality of possible meanings associated with the query based on the filtered plurality of pairs and the third term.

8. The method of claim 7, further comprising:

identifying, in the knowledge graph, a third candidate component associated with the third term;

identifying expected terms associated with the third candidate component;

comparing the expected terms associated with the third candidate component to each pair in the filtered plurality of pairs;

filtering out a subset of the plurality of pairs based on the comparison; and

determining the plurality of possible meanings associated with the query based on the subset of the plurality of pairs and the expected terms.

9. The method of claim 1 wherein selecting the meaning for the query further comprises:

generating for display the plurality of possible meanings associated with the query; and

receiving a user selection of the meaning for the query.

10. The method of claim 1 wherein selecting the meaning for the query further comprises:

ranking the plurality of possible meanings according to a user preference; and

automatically determining the meaning for the query based on the ranking.

11. A system for identifying a meaning of an ambiguous term in a natural language query, the system comprising:

a storage device;

a user interface; and

processing circuitry configured to;

receive, via the user interface, a query from a user;

isolate a first term and a second term from the query received from the user;

access, at a storage device, a knowledge graph;

identify, in the knowledge graph, a first plurality of candidate components associated with the first term, wherein the first term having more than one candidate component indicates the first term has an ambiguous meaning;

identify, in the knowledge graph, a second plurality of identified candidate components associated with the second term, wherein the second term having more than one candidate component indicates the second term has an ambiguous meaning;

match each candidate component of the first plurality of candidate components to each candidate component of the second plurality of candidate components to form a plurality of pairs;

determine a strength of association for each pair in the plurality of pairs based on a distance on the knowledge graph between two nodes of the respective pair;

filter the plurality of pairs by the strength of association for each pair;

determine a plurality of possible meanings associated with the query based on the filtered plurality of pairs; and

select a meaning from the plurality of possible meanings associated with the query based at least in part on a ranking of the plurality of possible meanings.

12. The system of claim 11, wherein the processing circuitry is further configured to:

isolate a command term from the query received from the user;

determine the plurality of possible meanings associated with the query based on the filtered plurality of pairs and the command term.

13. The system of claim 12, wherein the processing circuitry configured to determine the plurality of possible meanings is further configured to:

determine a plurality of possible meanings associated with the command term, wherein each of the possible meanings is associated with expected terms;

compare the expected terms associated with each possible meaning associated with the command term to the filtered plurality of pairs;

filter out a subset of the plurality of pairs based on the comparison; and

determine the plurality of possible meanings associated with the query based on the subset of the plurality of pairs and the expected terms.

14. The system of claim 11, wherein the processing circuitry configured to filter the plurality of pairs is further configured to:

identify a first pair of the plurality of pairs;

determine whether the first pair has a strength of association which is below a required threshold;

in response to determining that the strength of association is below the required threshold, exclude the first pair from the filtered plurality of pairs; and

in response to determining that the strength of association is not below the required threshold, include the first pair in the filtered the plurality of pairs.

15. The system of claim 14, wherein the processing circuitry is further configured to:

determine the required threshold based on the strength of association for each of the plurality of pairs;

adjust the required threshold based on the filtered plurality of pairs; and

filter the plurality of pairs based on the adjusted required threshold.

16. The system of claim 14, further comprising:

receive feedback on the accuracy for the plurality of possible meanings associated with the query; and

adjust the required threshold based on the received feedback.

17. The system of claim 11, wherein the processing circuitry is further configured to:

isolate a third term from the query received from the user;

determine the plurality of possible meanings associated with the query based on the filtered plurality of pairs and the third term.

18. The system of claim 17, wherein the processing circuitry is further configured to:

identify, in the knowledge graph, a third candidate component associated with the third term;

identify expected terms associated with the third candidate component;

compare the expected terms associated with the third candidate component to each pair in the filtered plurality of pairs;

filter out a subset of the plurality of pairs based on the comparison; and

determine the plurality of possible meanings associated with the query based on the subset of the plurality of pairs and the expected terms.

19. The system of claim 11 wherein the processing circuitry configured to select the meaning for the query is further configured to:

generate for display the plurality of possible meanings associated with the query; and

receive a user selection of the meaning for the query.

20. The system of claim 11 wherein the processing circuitry configured to select the meaning for the query is further configured to:

rank the plurality of possible meanings according to a user preference; and

automatically determine the meaning for the query based on the ranking.