Identifying subjective attributes by analysis of curation signals

Patent metadata

Field	Value
Publication number	US-9811780-B1
Application number	US-201313842896-A
Country	US
Kind code	B1
Filing date	Mar 15, 2013
Priority date	Aug 28, 2012
Publication date	Nov 7, 2017
Grant date	Nov 7, 2017

Abstract

A system and method for identifying and predicting subjective attributes for entities (e.g., media clips, images, newspaper articles, blog entries, persons, organizations, commercial businesses, etc.) are disclosed. In one aspect, a first set of subjective attributes for a first entity is identified based on a reaction to the first entity. A classifier is trained on a set of input-output mappings, wherein the set of input-output mappings comprises an input-output mapping whose input is based on a feature vector for the first entity and whose output is based on the first set of subjective attributes. A feature vector for a second entity is then provided to the trained classifier to obtain a second set of subjective attributes for the second entity.

First claim

What is claimed is:

1. A method comprising:
determining, by a computer system, a plurality of subjective attributes for a first entity based on terms from a plurality of user reactions to the first entity, wherein the terms are not a user-supplied tag;
determining respective relevancy scores for the plurality of subjective attributes with respect to the first entity using at least a number of occurrences of a plurality of users bookmarking the first entity;
obtaining a first set of features for the first entity; and
training a classifier on a set of input-output mappings, wherein the set of input-output mappings comprises an input-output mapping whose input is based on the first set of features and whose output is based on the respective relevancy scores for the plurality of subjective attributes with respect to the first entity, wherein the trained classifier is to be applied to a second set of features of a second entity to obtain respective relevancy scores for one or more subjective attributes with respect to the second entity.

2. The method of claim 1 wherein the determining of the plurality of subjective attributes for the first entity is further based on a plurality of users sharing the first entity.

3. The method of claim 1 wherein the determining of the plurality of subjective attributes for the first entity is further based on a plurality of users adding the first entity to respective playlists.

4. The method of claim 1 wherein the first entity is a media clip.

5. The method of claim 1 wherein the respective relevancy score for a subjective attribute is based on a number of times that the subjective attribute appears in a set of comments pertaining to the first entity.

6. The method of claim 1 wherein the respective relevancy score for a subjective attribute is based on a number of comments pertaining to the first entity.

7. An apparatus comprising:
a memory to store:
one or both of a first entity and a datum associated with the first entity, and
one or both of a second entity and a datum associated with the second entity; and
a processor, operatively coupled to the memory, to:
determine a plurality of subjective attributes for a first entity based on terms from a plurality of user reactions to the first entity, wherein the terms are not a user-supplied tag;
determine respective relevancy scores for the plurality of subjective attributes with respect to the first entity using at least a number of occurrences of a plurality of users bookmarking the first entity;
obtain a first set of features for the first entity; and
train a classifier on a set of input-output mappings, wherein the set of input-output mappings comprises an input-output mapping whose input is based on the first set of features and whose output is based on the respective relevancy scores for the plurality of subjective attributes with respect to the first entity, wherein the trained classifier is to be applied to a second set of features of a second entity to obtain respective relevancy scores for one or more subjective attributes with respect to the second entity.

8. The apparatus of claim 7 wherein the determining is further based on a plurality of users adding the first entity to respective playlists.

9. The apparatus of claim 7 wherein the first entity is a person or an organization.

10. The apparatus of claim 7 wherein the processor is further to provide a suggestion that suggests one or more subjective attributes for a third entity.

11. The apparatus of claim 10 wherein the processor is further to re-train the classifier based on a response to the suggestion.

12. A non-transitory computer-readable storage medium having instructions stored therein, which when executed, cause a computer system to perform operations comprising:
determining a plurality of subjective attributes for a first entity based on terms from a plurality of user reactions to the first entity, wherein the terms are not a user-supplied tag;
determining respective relevancy scores for the plurality of subjective attributes with respect to the first entity using at least a number of occurrences of a plurality of users bookmarking the first entity;
obtaining a first set of features for the first entity; and
training a classifier on a set of input-output mappings, wherein the set of input-output mappings comprises an input-output mapping whose input is based on the first set of features and whose output is based on the respective relevancy scores for the plurality of subjective attributes with respect to the first entity, wherein the trained classifier is to be applied to a second set of features of a second entity to obtain respective relevancy scores for one or more subjective attributes with respect to the second entity.

13. The non-transitory computer-readable storage medium of claim 12 wherein the first entity is a blog entry.

14. The non-transitory computer-readable storage medium of claim 12 wherein the operations further comprise obtaining the data pertaining to the one or more users bookmarking the one or more entities.

15. The non-transitory computer-readable storage medium of claim 14 wherein the set of input-output mappings comprises data pertaining to a source of a comment that mentions the subjective attribute with respect to the first entity.

16. The non-transitory computer-readable storage medium of claim 12 wherein the operations further comprise selecting subjective attributes with the highest relevancy scores.

17. The non-transitory computer-readable storage medium of claim 12 wherein the operations further comprise applying a threshold to the respective relevancy scores.

Assignees

• Aradhye Hrishikesh
• Shetty Sanketh
• Google Inc

Inventors

• Aradhye Hrishikesh
• Shetty Sanketh

Classifications

• G06N99/005 Primary

  Physics · mapped topic

• G06F17/3053

  Physics · mapped topic

• G06N20/00 Primary

  Machine learning · CPC title

• G06F16/24578

  using ranking · CPC title