Method and system for capturing and exploiting user intent in a conversational interaction based information retrieval system

Patent metadata

Field	Value
Publication number	US-9946757-B2
Application number	US-201414274147-A
Country	US
Kind code	B2
Filing date	May 9, 2014
Priority date	May 10, 2013
Publication date	Apr 17, 2018
Grant date	Apr 17, 2018

Abstract

Systems and methods are disclosed for processing a search request received from a user. A method includes providing access to content items associated with metadata. The method includes providing information about at least one search previously performed for a user, and providing access to information describing relationships between at least one of the content items and the metadata. The method includes receiving a present input from the user intended to identify a desired content item. Upon determining that results from the previous search did not satisfy the search request from the user, the method includes determining an intent shared by the previous search and the present input based on comparing (i) the relationship information, (ii) the previous search, and (iii) the present input. The method also includes selecting and presenting a subset of content items based on comparing the shared intent and the metadata.

First claim

What is claimed is:

1. A computer-implemented method of processing a search request received from a user, the method comprising:
accessing a database comprising a set of content items and metadata that describes corresponding content items in the set of content items;
receiving at least one previous search from a user, wherein the at least one previous search comprises a first set of media assets and a window of time;
determining that results from the at least one previous search did not satisfy the user based on determining that a time period of availability of the first set of media assets is outside the window of time;
receiving a present input from the user, wherein the present input comprises a second set of media assets;
determining that results from the present input will not satisfy the user based on determining that a time period of availability of the second set of media assets is outside the window of time; and
based on determining that the results from the at least one previous search did not satisfy the user and based on determining that the results from the present input will not satisfy the user:
determining an intent shared by the at least one previous search and the present input based on:
determining a first category that describes a first subset of the set of content items, wherein the first subset includes the first set of media assets and the second set of media assets;
determining a second category that describes a second subset of the set of content items, wherein the second subset includes the first set of media assets and the second set of media assets;
identifying at least one of a previous smart tag, a previous category, and a previous microcategory associated with the at least one previous search;
identifying at least one of a present smart tag, a present category, and a present microcategory associated with the present input; and
determining the intent based on determining at least one relationship between at least one of (a) the previous smart tag and the present smart tag, (b) the previous category and the present category, (c) the previous microcategory and the present microcategory, and (d) a first number of content items in the first subset of content items and a second number of content items in the second subset of content items;
searching the database for a third set of media assets consistent with the determined intent and having a time period of availability within the window of time; and
presenting the third set of media assets to the user.

2. The method of claim 1 , wherein the determining the at least one relationship is based on determining a measure of relatedness based on a collection of relationship distances of entities,
wherein the entities include at least one of (a) a content item, (b) an element of the at least one previous search, and (c) an element of the present input, and
wherein the relatedness measure is based on one of the relationship distances.

3. The method of claim 2 , wherein the collection of relationship distances of entities includes a relationship graph with nodes and edges,
the nodes representing the entities, and
the edges representing a direct relationship between any two entities.

4. The method of claim 3 , wherein the relatedness measure of two entities is measured in a number of hops between two nodes corresponding to the two entities.

5. The method of claim 3 , wherein each edge has a weight, wherein the relationship distance of two entities is a combination of weights of edges connecting two nodes corresponding to the two entities.

6. The method of claim 2 , wherein the relationship distances are modified by user preferences acquired over time.

7. The method of claim 2 , wherein the determining the at least one relationship is based further on a collection of historical inputs, wherein the relationship distances of the entities are determined based on how often the entities were used together according to the historical inputs.

8. The method of claim 2 , wherein the determining the at least one relationship is based further on user preferences acquired over time, and wherein the relatedness measure is modified based on the user preferences.

9. A system for processing a search request received from a user, the system comprising:
computer readable instructions encoded on a non-transitory computer readable medium, the computer readable instructions causing the computer system to:
access a database comprising a set of content items and metadata that describes corresponding content items in the set of content items;
receive at least one previous search from a user, wherein the at least one previous search comprises a first set of media assets and a window of time;
determine that results from the at least one previous search did not satisfy the user based on determining that a time period of availability of the first set of media assets is outside the window of time;
receive a present input from the user, wherein the present input comprises a second set of media assets;
determine that results from the present input will not satisfy the user based on determining that a time period of availability of the second set of media assets is outside the window of time;
based on determining that the results from the at least one previous search did not satisfy the user and based on determining that the results from the present input will not satisfy the user:
determine an intent shared by the at least one previous search and the present input based on:
determining a first category that describes a first subset of the set of content items, wherein the first subset includes the first set of media assets and the second set of media assets;
determining a second category that describes a second subset of the set of content items, wherein the second subset includes the first set of media assets and the second set of media assets;
identifying at least one of a previous smart tag, a previous category, and a previous microcategory associated with the at least one previous search;
identifying at least one of a present smart tag, a present category, and a present microcategory associated with the present input;
determining the intent based on a determination of at least one relationship between at least one of (a) the previous smart tag and the present smart tag, (b) the previous category and the present category, (c) the previous microcategory and the present microcategory and (d) a first number of content items in the first subset of content items and a second number of content items in the second subset of content items;
search the database for a third set of media assets consistent with the determined intent and having a time period of availability within the window of time; and
present the third set of media assets to the user.

10. The system of claim 9 , wherein the determination of the at least one relationship is based on determining a measure of relatedness based on a collection of relationship distances of entities,
wherein the entities include at least one of (a) a content item, (b) an element of the at least one previous search, and (c) an element of the present input, and
wherein the relatedness measure is based on one of the relationship distances.

11. The system of claim 10 , wherein the collection of relationship distances of entities includes a relationship graph with nodes and edges,
the nodes representing the entities, and
the edges representing a direct relationship between any two entities.

12. The system of claim 11 , wherein the relatedness measure of two entities is measured in a number of hops between two nodes corresponding to the two entities.

Assignees

• Veveo Inc

Inventors

• Aravamudan Murali
• Venkataraman Sashikumar
• Barve Rakesh
• Mohaideen Pathurudeen Ahmed Nizam

Classifications

• G06F17/30861

  Physics · mapped topic

• G06F17/30522 Primary

  Physics · mapped topic

• G06F16/2457 Primary

  with adaptation to user needs · CPC title

• G06F16/95 Primary

  Retrieval from the web · CPC title