Customizing digital content with editorial content

Patent metadata

Field	Value
Publication number	US-11720616-B2
Application number	US-202117153321-A
Country	US
Kind code	B2
Filing date	Jan 20, 2021
Priority date	Feb 6, 2017
Publication date	Aug 8, 2023
Grant date	Aug 8, 2023

Abstract

A digital magazine server generates a digital magazine for user based on a received request for the digital magazine identifying one or more topics. The digital magazine server applies one or more machined trained models to obtained content items to select content items for the topic. A hierarchy of the topics included in the received request may be determined by the digital magazine server and used by the trained models to select content items. When generating the digital magazine, the digital magazine server also includes one or more editorial content items that are manually selected. The digital magazine serer may reposition one or more content items selected by the trained models to include an editorial content items.

First claim

What is claimed is:

1. A method, comprising:
receiving a request from a user for a digital magazine at a digital magazine server, wherein the request comprises a plurality of topics;
determining a hierarchy of the plurality of topics identified by the request, the hierarchy identifying a root topic and one or more sub-topics of the root topic from the plurality of topics, each of the one or more sub-topics being a subset of the root topic;
selecting one or more content items from a plurality of content items to include in the digital magazine requested by the user, wherein the one or more content items are selected based on the determined hierarchy of the plurality of topics and a trained model, the model trained to output a selection of content items from a pool of content items based on topics;
generating the digital magazine by combining the one or more content items with a manually selected content item, wherein combining the one or more content items with the manually selected content item comprises:
determining a positional index of the manually selected content item based on a topic of the manually selected content item that corresponds to a topic of determined hierarchy of the plurality of topics, the positional index specifying a location where the manually selected content item is to be presented relative to the one or more content items;

transmitting the generated digital magazine to a client device for presentation to the user, wherein the generated digital magazine is presented such that the manually selected content item is presented relative to the one or more content items in accordance with the positional index.

2. The method of claim 1 , wherein selecting the one or more content items from the plurality of content items based at least on the hierarchy of the plurality of topics and the trained model comprises:
applying a logical AND operation to the root topic and the one or more sub-topics;
applying the trained model to a result of the logical operation and the plurality of content items; and
receiving a selection of a subset of the plurality of content items from the trained model.

3. The method of claim 2 , further comprising:
identifying the one or more content items from the subset based on sources of the one or more content items.

4. The method of claim 1 , wherein generating the digital magazine by combining the one or more content items in the subset with the manually selected content items comprises:
ranking the one or more content items based on prior interaction with the one or more content items by users of the digital magazine server; and
generating the digital magazine based on the ranking.

5. The method of claim 4 , wherein the prior interaction with the one or more candidate content items by users of the digital magazine server is selected from a group consisting of: sharing a candidate content item with another user, indicating a preference for the candidate content item, indicating a dislike for the candidate content item, including the candidate content item in a digital magazine, storing the candidate content item, commenting on the candidate content item, accessing the candidate content item, and any combination thereof.

6. The method of claim 1 , further comprising:
extracting a new feature from the root topic and the one or more sub-topics identified from the request; and
using the new feature to further train the trained model.

7. A non-transitory computer readable medium storing executable computer program instructions, the computer program instructions comprising instructions that when executed cause a computer processor to:
receive a request from a user for a digital magazine at a digital magazine server, wherein the request comprises a plurality of topics;
determine a hierarchy of the plurality of topics identified by the request, the hierarchy identifying a root topic and one or more sub-topics of the root topic from the plurality of topics, each of the one or more sub-topics being a subset of the root topic;
select one or more content items from a plurality of content items to include in the digital magazine requested by the user, wherein the one or more content items are selected based on the determined hierarchy of the plurality of topics and a trained model, the model trained to output a selection of content items from a pool of content items based on topics;
generate the digital magazine by combining the one or more content items with a manually selected content item, wherein combining the one or more content items with the manually selected content item comprises:
determining a positional index of the manually selected content item based on a topic of the manually selected content item that corresponds to a topic of determined hierarchy of the plurality of topics, the positional index specifying a location where the manually selected content item is to be presented relative to the one or more content items; and

transmit the generated digital magazine to a client device for presentation to the user, wherein the generated digital magazine is presented such that the manually selected content item is presented relative to the one or more content items in accordance with the positional index.

8. The non-transitory computer readable medium of claim 7 , wherein the computer program instructions for selecting the one or more content items from the plurality of content items based at least on the hierarchy of the plurality of topics and the trained model comprise instructions that when executed cause the computer processor to:
apply a logical AND operation to the root topic and the one or more sub-topics;
apply the trained model to a result of the logical operation and the plurality of content items; and
receiving a selection of a subset of the plurality of content items from the trained model.

9. The non-transitory computer readable medium of claim 8 , wherein the computer program instructions further comprise instructions that when executed cause a computer processor to:
identify the one or more content items from the subset based on sources of the one or more content items.

10. The non-transitory computer readable medium of claim 7 , wherein the computer program instructions for generating the digital magazine by combining the one or more content items in the subset with the manually selected content items comprise instructions that when executed cause the computer processor to:
rank the one or more content items based on prior interaction with the one or more content items by users of the digital magazine server; and
generate the digital magazine based on the ranking.

11. The non-transitory computer readable medium of claim 10 , wherein the prior interaction with the one or more candidate content items by users of the digital magazine server is selected from a group consisting of: sharing a candidate content item with another user, indicating a preference for the candidate content item, indicating a dislike for the candidate content item, including the candidate content item in a digital magazine, storing the candidate content item, commenting on the candidate content item, accessing the candidate content item, and any combination thereof.

12. The non-transitory computer readable medium of claim 7 , wherein the computer program instructions further comprise instructions that when executed cause a computer processor to:
extract a new feature from the root topic and the one or more sub-topics identified from the request; and
use the new feature to further train the trained model.

13. A computer system, comprising:
a computer processor for executing computer program instruction

Assignees

• Flipboard Inc

Inventors

• Cora Vlad Ionut
• Frederickson Benjamin John
• Mazzeo John T
• Mccue Michael S

Classifications

• G06F16/34 Primary

  Browsing; Visualisation therefor (browsing or visualisation for clustering or classification G06F16/358) · CPC title

• G06F16/81

  Indexing, e.g. XML tags; Data structures therefor; Storage structures · CPC title

• G06F16/9577

  Optimising the visualization of content, e.g. distillation of HTML documents · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06N20/00

  Machine learning · CPC title