Systems and methods for determining alternative names
View Patent ↗Systems and methods described herein provide for determining alternative names for a target name. A plurality of candidate alternative names is retrieved. For example, if the target name is “bush”, the candidate alternative names may include “George W. Bush”, “George H. W. Bush”, “Jeb Bush”, “Bush, Illinois (location)”, and “Bush (plant)”. A plurality of information sources associated with the target name are identified (e.g., documents, media assets or other information associated with “bush”). A first value and a second value are computed for a first candidate name (e.g., “George W. Bush”) and second candidate name (“George H. W. Bush”), respectively, based on how many of the identified information sources refer to the first and second candidate names, respectively. In response to determining that the first value exceeds the second value, the first candidate name, instead of the second candidate name, is assigned as an alternative name for the target name.
1. A method for determining alternative names, the method comprising:
retrieving, based on input from a user, a plurality of candidate alternative names for a target name;
storing the plurality of candidate alternative names;
identifying a plurality of media assets associated with the target name, wherein each of the plurality of media assets is provided by a source other than the user;
generating a first classification group that includes each of the plurality of media assets that refers to a first candidate name;
generating a second classification group that includes each of the plurality of media assets that refers to a second candidate name;
computing a first value representing a first number of media assets in the first classification group;
computing a second value representing a second number of media assets in the second classification group;
in response to determining that the first value exceeds the second value, assigning the first candidate name, instead of the second candidate name, as an alternative name for the target name; and
in response to determining that the second value exceeds the first value, assigning the second candidate name, instead of the first candidate name, as the alternative name for the target name.
2. The method of claim 1 , wherein:
the first value is a first proportion of the identified plurality of media assets that refers to the first candidate name, and
the second value is a second proportion of the identified plurality of media assets that refers to the second candidate name.
3. The method of claim 1 , wherein:
the first value is computed based on a first metadata associated with the first candidate name, and
the second value is computed based on a second metadata associated with the second candidate name, wherein
the first metadata and the second metadata are each retrieved from at least one of the plurality of media assets.
4. The method of claim 3 , wherein the first metadata indicates at least one of:
a position of the first candidate name in a media asset of the plurality of media assets, and
a hyperlink associated with the first candidate name in a media asset of the plurality of media assets, wherein the hyperlink is associated with a higher weight than the position in computing the first value.
5. The method of claim 1 , wherein each of the plurality of media assets is one of:
a webpage associated with the target name,
an article associated with the target name,
a book associated with the target name, and
a media asset associated with the target name.
6. The method of claim 1 , further comprising:
retrieving, using the control circuitry, a list of disambiguation documents associated with the target name, wherein the plurality of candidate alternative names is extracted from the retrieved list of disambiguation documents.
7. The method of claim 1 , wherein the target name is associated with a media asset, and wherein each of the plurality of candidate alternative names represents a different candidate category for the media asset.
8. The method of claim 1 , wherein the target name is associated with a target information source, further comprising:
filtering, using the control circuitry, the plurality of candidate alternative names, wherein the filtering comprises removing one of the plurality of candidate alternative names in response to determining that the target information source does not refer to the one of the plurality of candidate names.
9. The method of claim 1 , wherein the target name is a primary keyword in a plurality of keywords associated with a media asset, further comprising:
identifying a secondary keyword from the plurality of keywords;
computing, using the control circuitry, the first value based on a first correlation value between the first candidate name and the secondary keyword; and
computing, using the control circuitry, the second value based on a second correlation value between the second candidate name and the secondary keyword.
10. The method of claim 1 , further comprising:
computing a first correlation value based on the number of times a secondary keyword appears in a first media asset associated with the first candidate name; and
computing a second correlation value based on the number of times the secondary keyword appears in a second media asset associated with the second candidate name.
11. A system for determining alternative names, the system comprising:
storage circuitry configured to:
store a plurality of candidate alternative names for a target name, wherein the plurality of candidate alternative names;
control circuitry configured to:
retrieve, based on input from a user, the plurality of candidate names;
identify a plurality of media assets associated with the target name, wherein each of the plurality of media assets is provided by a source other than the user;
generate a first classification group that includes each of the plurality of media assets that refers to a first candidate name;
generate a second classification group that includes each of the plurality of media assets that refers to a second candidate name;
compute a first value representing a first number of media assets in the first classification group;
compute a second value representing a second number of media assets in the second classification group;
assign the first candidate name, instead of the second candidate name, as an alternative name for the target name responsive to a determination that the first value exceeds the second value; and
assign the second candidate name, instead of the first candidate name, as the alternative name for the target name responsive to a determination that the second value exceeds the first value.
12. The system of claim 11 , wherein:
the first value is a first proportion of the identified plurality of media assets that refers to the first candidate name, and
the second value is a second proportion of the identified plurality of media assets that refers to the second candidate name.
13. The system of claim 11 , wherein:
the first value is computed based on a first metadata associated with the first candidate name, and
the second value is computed based on a second metadata associated with the second candidate name, wherein
the first metadata and the second metadata are each retrieved from at least one of the plurality of media assets.
14. The system of claim 13 , wherein the first metadata indicates at least one of:
a position of the first candidate name in a media asset of the plurality of media assets, and
a hyperlink associated with the first candidate name in a media asset of the plurality of media assets, wherein the hyperlink is associated with a higher weight than the position in computing the first value.
15. The system of claim 11 , wherein each of the plurality of media assets is one of:
a webpage associated with the target name,
an article associated with the target name,
a book associated with the target name, and
a media asset associated with the target name.
16. The system of claim 11 , wherein the control circuitry is further configured to:
retrieve a list of disambiguation documents associated with the target name, wherein the plurality of candidate alternative names is extracted from the retrieved list of disambiguation documents.
17. The system of claim 11 , wherein the target name is associated with a media asset, and wherein each of the plurality of candidate alternative names represents a different candidate category for the media asset.
18. The system of claim 11 , wherein the target name is associated with a target information source, wherein the control circuitry is further configured to:
filter the plurality of candidate alternative names, wherein the filtering comprises removing one of from the plurality of candidate alternative names in response to determining that the target information source does not refer to the one of the plurality of candidate names.
19. The system of claim 11 , wherein the target name is a primary keyword in a plurality of keywords associated with a media asset, wherein the control circuitry is further configured to:
identify a secondary keyword from the plurality of keywords;
compute the first value based on a first correlation value between the first candidate name and the secondary keyword; and
compute the second value based on a second correlation value between the second candidate name and the secondary keyword.
20. The system of claim 11 , wherein the control circuitry is further configured to:
compute a first correlation value based on the number of times a secondary keyword appears in a first media asset associated with the first candidate name; and
compute a second correlation value based on the number of times the secondary keyword appears in a second media asset associated with the second candidate name.