Disambiguating user intent in conversational interaction system for large corpus information retrieval

• US 10,572,520 B2
• Filed: 09/27/2016
• Issued: 02/25/2020
• Est. Priority Date: 07/31/2012
• Status: Active Grant

• Current Assignee: Veveo, Inc. (Adeia Inc.)
• Original Assignee: Veveo, Inc. (Adeia Inc.)
• Inventors: Aravamudan, Murali, Barve, Rakesh, Venkataraman, Sashikumar, Agarwal, Vineet, Savarkar, Abhijit, Ramamoorthy, Ganesh, Mohaideen Pathurudeen, Ahmed Nizam, Reddy, K China Srinivasulu
• Primary Examiner(s): Beausoliel, Jr., Robert W
• Assistant Examiner(s): Santos, Pedro J
• Application Number:
  US15/277,288
• Publication Number:
  US 20170017719A1
• Time in Patent Office: 1,246 Days
• Field of Search: 707999003, 707E17108, 707748, 707E17109, 707E17102, 707741, 707752
• CPC Class Codes: G06F 16/24 Querying
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/3328 using graphical result space presentation or visualisation
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G06F 16/367 Ontology
• CPC Class Codes: G06F 16/90324 using system suggestions

Abstract

A method of disambiguating user intent in conversational interactions for information retrieval is disclosed. The method includes providing access to a set of content items with metadata describing the content items and providing access to structural knowledge showing semantic relationships and links among the content items. The method further includes providing a user preference signature, receiving a first input from the user that is intended by the user to identify at least one desired content item, and determining an ambiguity index of the first input. If the ambiguity index is high, the method determines a query input based on the first input and at least one of the structural knowledge, the user preference signature, a location of the user, and the time of the first input and selects a content item based on comparing the query input and the metadata associated with the content item.

Claims

1. A method of disambiguating user intent in conversational interactions for information retrieval, the method comprising:

receiving information for a set of content items, each of the set of content items being associated with metadata that describes the corresponding content item;

receiving information for structural knowledge, the structural knowledge showing semantic links between content items;

receiving information for a user preference signature, the user preference signature describing preferences of a user for at least one of (i) particular content items and (ii) metadata associated with each of the particular content items;

receiving a first input from the user, the first input intended by the user to identify at least one desired content item and including at least one word;

determining an ambiguity level of the at least one word;

determining whether the ambiguity level exceeds a first threshold value;

in response to determining that the ambiguity level exceeds the first threshold value;

determining a disambiguated first input based on the first input and at least one of the structural knowledge, the user preference signature, a location of the user, and a time of the first input; and

selecting a subset of content items from the set of content items based on comparing the disambiguated first input with the metadata associated with each of the set of content items; and

in response to determining that the ambiguity level does not exceed the first threshold value;

selecting a subset of content items from the set of content items based on comparing the first input with the metadata associated with each of the set of content items.

2. The method of claim 1, wherein the ambiguity level exceeds the first threshold value, further comprising:

upon a condition in which the at least one of the structural knowledge, the user preference signature, the location of the user, and the time of the first input does not disambiguate the first input;

presenting a disambiguating question to the user;

receiving a second input from the user in response to the disambiguating question; and

determining the disambiguated first input further based on the second input.

3. The method of claim 2, wherein the disambiguating question presents to the user a plurality of options for the first input.

4. The method of claim 1, wherein the ambiguity level exceeds the first threshold value, further comprising:

upon a condition in which the at least one of the structural knowledge, the user preference signature, the location of the user, and the time of the first input does not disambiguate the first input;

generating for display a notification for the user that multiple options exist for the first input; and

selecting the multiple options as the subset of content items.

5. The method of claim 1, wherein determining the disambiguated first input further comprises:

determining the disambiguated first input based on the user preference signature upon a condition in which the user preference signature is available; and

determining the disambiguated first input based on at least one of the structural knowledge, the location of the user, and the time of the first input upon a condition in which the user preference signature is not available.

6. The method of claim 1, wherein determining the disambiguated first input is further based on user history.

7. The method of claim 1, wherein the subset of content items comprises episodes of a series, and wherein selecting the subset of content items comprises selecting a second episode of the series that follows a first episode of the series that the user most recently watched.

8. The method of claim 1, wherein the subset of content items comprises at least one location, and further comprising displaying the at least one location on a map.

9. The method of claim 1, wherein selecting the subset of content items from the set of content items based on comparing the disambiguated first input with the metadata associated with each of the set of content items comprises:

determining, based on the comparing, that no content items in the set of content items satisfy the disambiguated first input; and

selecting a subset of content items that satisfies a portion of the disambiguated first input.

10. A system for disambiguating user intent in conversational interactions for information retrieval, the system comprising a processor configured to:

receive information for a set of content items, each of the set of content items being associated with metadata that describes the corresponding content item;

receive information for structural knowledge, the structural knowledge showing semantic links between content items;

receive information for a user preference signature, the user preference signature describing preferences of a user for at least one of (i) particular content items and (ii) metadata associated with each of the particular content items;

receive a first input from the user, the first input intended by the user to identify at least one desired content item and including at least one word;

determine an ambiguity level of the at least one word;

determine whether the ambiguity level exceeds a first threshold value;

in response to determining that the ambiguity level exceeds the first threshold value;

determine a disambiguated first input based on the first input and at least one of the structural knowledge, the user preference signature, a location of the user, and a time of the first input; and

select a subset of content items from the set of content items based on comparing the disambiguated first input with the metadata associated with each of the set of content items; and

in response to determining that the ambiguity level does not exceed the first threshold value;

select a subset of content items from the set of content items based on comparing the first input with the metadata associated with each of the set of content items.

11. The system of claim 10, wherein the ambiguity level exceeds the first threshold value, and wherein the processor is further configured to:

upon a condition in which the at least one of the structural knowledge, the user preference signature, the location of the user, and the time of the first input does not disambiguate the first input;

present a disambiguating question to the user;

receive a second input from the user in response to the disambiguating question; and

determine the disambiguated first input further based on the second input.

12. The system of claim 11, wherein the disambiguating question presents to the user a plurality of options for the first input.

13. The system of claim 10, wherein the ambiguity level exceeds the first threshold value, and wherein the processor is further configured to:

upon a condition in which the at least one of the structural knowledge, the user preference signature, the location of the user, and the time of the first input does not disambiguate the first input;

generate for display a notification for the user that multiple options exist for the first input; and

select the multiple options as the subset of content items.

14. The system of claim 10, wherein the processor is further configured, when determining the disambiguated first input, to:

determine the disambiguated first input based on the user preference signature upon a condition in which the user preference signature is available; and

determine the disambiguated first input based on at least one of the structural knowledge, the location of the user, and the time of the first input upon a condition in which the user preference signature is not available.

15. The system of claim 10, wherein the processor is configured, when determining the disambiguated first input, to determine the disambiguated first input further based on user history.

16. The system of claim 10, wherein the subset of content items comprises episodes of a series, and wherein the processor is further configured, when selecting the subset of content items, to select a second episode of the series that follows a first episode of the series that the user most recently watched.

17. The system of claim 10, wherein the subset of content items comprises at least one location, and wherein the processor is further configured to display the at least one location on a map.

18. The system of claim 10, wherein the processor is further configured, when selecting the subset of content items from the set of content items based on comparing the disambiguated first input with the metadata associated with each of the set of content items, to:

determine, based on the comparing, that no content items in the set of content items satisfy the disambiguated first input; and

select a subset of content items that satisfies a portion of the disambiguated first input.