Identifying Similar Content on a Digital Magazine Server

• US 20150100587A1
• Filed: 10/08/2013
• Published: 04/09/2015
• Est. Priority Date: 10/08/2013
• Status: Active Grant

• Current Assignee: Flipboard Incorporated
• Original Assignee: Flipboard Incorporated
• Inventors: Walkingshaw, Andrew David, Fei, Xiaoyu
• Granted Patent: US 10,437,901 B2
• Time in Patent Office: Days
• US Class Current: 707/748
• CPC Class Codes: G06F 16/954 Navigation, e.g. using categorised browsing

Abstract

A digital magazine identifies content items for recommendation to a user by determining the address information of one or more content items. The address information of a content item identifies a source from which the content item was received and additional information describing the content item. Representations of content items are generated from the address information for each content item. Additionally, representations of content items previously presented to the user are generated from the address information for the previously presented content items. A measure of similarity between a representation of a content item and a representation of a previously presented content item is determined and used to select content items to present to the user.

Claims

1. A method for identifying content items for presentation to a user of a digital magazine server, the method comprising:

retrieving one or more content items from one or more sources;

determining address information for each of the retrieved one or more content items, the address information for a content item specifying a source from which the content item was retrieved and information describing the content item;

identifying one or more attributes associated with each of the retrieved content items, one or more attributes of the content item determined from the address information for the content item;

generating representations of the address information for each of the content items based at least in part on one or more attributes identified from the address information;

determining representations of one or more content items previously presented to the user by the digital magazine server, each representation of a content item previously presented to the user based at least in part on the address information for the content item previously presented to the user;

determining a measure of similarity between one or more of the content items and the one or more content items previously presented to the user based at least in part on the representations of the address information for each of the one or more content items and the representations of the address information for the one or more content items previously presented to the user; and

selecting, for presentation to the user, at least one of the one or more content items with at least a threshold measure of similarity with one or more of the content items previously presented to the user.

2. The method of claim 1, wherein generating representations of the address information for each of the content items based at least in part on one or more attributes identified from the address information comprises:

assigning one or more weights to the one or more attributes identified from the address information for one or more content items; and

generating a representation of the address information for the content item based at least in part on values of one or more attributes identified from the address information for the content item and weights associated with one or more of the attributes identified from the address information for the content item.

3. The method of claim 2, wherein one or more of the weights are assigned based at least in part on one or more selected from a group consisting of:

information in a user profile associated with the user, one or more attributes of content items with which the user previously interacted, and any combination thereof.

4. The method of claim 1, wherein generating representations of the address information for each of the content items based at least in part on one or more attributes identified from the address information comprises:

associating a weight with one or more of the representations of the address information for the content items.

5. The method of claim 4, wherein a weight is associated with a representation of the address information for the content item based at least in part on one or more selected from a group consisting of:

information in a user profile associated with the user, one or more attributes of content items with which the user previously interacted, and any combination thereof.

6. The method of claim 1, wherein a weight is associated with a representation of the address information for the content item based at least in part on one or more settings of the digital magazine server.

7. The method of claim 1, wherein the measure of similarity between a content item and a content item previously presented to the user is a cosine similarity between the representation of the content item and the representation of the content item previously presented to the user.

8. The method of claim 1, wherein a representation of the address information for the content item comprises a vector.

9. A method for identifying content items for presentation to a user of a digital magazine server, the method comprising:

retrieving one or more content items from one or more sources presented to the user;

retrieving additional one or more content items from one or more sources presented to an additional user of the digital magazine server;

determining address information for each of the retrieved one or more content items presented to the user, the address information for a content item presented to the user specifying a source from which the content item presented to the user was retrieved and information describing the content item presented to the user;

determining address information for each of the retrieved one or more additional content items presented to the additional user, the address information for an additional content item presented to the additional user specifying a source from which the additional content item presented to the additional user was retrieved and information describing the additional content item presented to the additional user;

identifying one or more attributes associated with each of the retrieved one or more content items presented to the user, one or more attributes of the content item presented to the user determined from the address information for the content item presented to the user;

identifying one or more attributes associated with each of the retrieved one or more additional content items presented to the additional user, one or more attributes of the additional content item presented to the additional user determined from the address information for the additional content item presented to the additional user;

generating representations of the address information for each of the one or more content items presented to the user based at least in part on one or more attributes identified from the address information for each of the one or more content items presented to the user;

generating a combined representation of the one or more content items presented to the user based at least in part on the generated representations of the address information for each of the one or more content items presented to the user;

generating additional representations of the address information for each of the one or more additional content items presented to the additional user based at least in part on one or more attributes identified from the address information for each of the one or more additional content items presented to the additional user;

generating an additional combined representation of the one or more additional content items presented to the additional user based at least in part on the generated additional representations of the address information for each of the one or more additional content items presented to the additional user;

determining a measure of similarity between the combined representation and the additional combined representation; and

identifying to the user the additional one or more content items presented to the additional user if the measure of similarity has at least a threshold value.

10. The method of claim 9, wherein generating representations of the address information for each of the one or more content items presented to the user comprises:

assigning one or more weights to the one or more attributes identified from the address information for the one or more content items presented to the user; and

generating a representation of the address information for the content item presented to the user based at least in part on values of one or more attributes identified from the address information for the content item presented to the user and weights associated with one or more of the attributes identified from the address information for the content item presented to the user.

11. The method of claim 9, wherein generating additional representations of the address information for each of the one or more additional content items presented to the additional user comprises:

assigning one or more weights to the one or more attributes identified from the address information for the one or more additional content items presented to the additional user; and

generating an additional representation of the address information for the additional content item presented to the additional user based at least in part on values of one or more attributes identified from the address information for the additional content item presented to the additional user and weights associated with one or more of the attributes identified from the address information for the additional content item presented to the additional user.

12. The method of claim 9, wherein determining the measure of similarity between the combined representation and the additional combined representation comprises:

determining a cosine similarity between the combined representation and the additional combined representation.

13. The method of claim 9, wherein generating the combined representation of the one or more content items comprises:

generating a centroid of the generated representations of the address information for each of the one or more content items presented to the user.

14. The method of claim 9, wherein generating the additional combined representation of the one or more additional content items presented to the additional user comprises:

generating a centroid of the generated additional representations of the additional address information for each of the one or more additional content items presented to the additional user.

15. The method of claim 9, wherein identifying to the user the additional one or more content items presented to the additional user if the measure of similarity has at least the threshold value comprises:

presenting the additional one or more content items to the user if the measure of similarity has at least the threshold value.

16. The method of claim 9, wherein identifying to the user the additional one or more content items presented to the additional user if the measure of similarity has at least the threshold value comprises:

presenting information identifying the additional user to the user if the measure of similarity has at least the threshold value.

17. A method for determining similarity between content items associated with a digital magazine server, the method comprising:

retrieving a content item from a source;

determining address information for the content item, the address information specifying the source and information describing the content item;

retrieving an additional content item from an additional source;

determining additional address information for the additional content item, the additional address information specifying the additional source and information describing the additional content item;

identifying one or more attributes associated with the content item from the address information for the content item;

identifying one or more attributes associated with the additional content item from the additional address information for the additional content item;

generating a representation of the address information for the content item based at least in part on one or more attributes identified from the address information for the content item;

generating an additional representation of the additional address information for the additional content item based at least in part on one or more attributes identified from the additional address information for the additional content item; and

determining a measure of similarity between the content item and the additional content item based at least in part on the representation of the address information for the content item and the representation of the additional address information for the additional content item.