Back to results
US 10,757,217 B2Granted
Determining viewer affinity for articles in a heterogeneous content feed
AssigneeMicrosoft Technology Licensing, LLC
InventorsAnkan Saha, Ajith Muralidharan
Life of this patent
Estimating the term…
Abstract
The present disclosure describes various embodiments of methods, systems, and machine-readable mediums which help determine a user's likely affinity for an article presented (or to be presented) in a heterogeneous feed of a social network.
Claims12
A method comprising:
- retrieving, by a computer system from a database in communication with the computer system, a member profile of a member of a social networking service, the member profile including information about the member, said information provided by the member;
- identifying, by the computer system, a plurality of features associated with the member based on the member profile;
- identifying, by the computer system, a plurality of features associated with an article eligible for sharing in a feed of the social networking service;
- determining, by the computer system, based on the plurality of features associated with the member and the plurality of features associated with the article, a scalar score reflecting a level of affinity of the member for the article;
- and
- for each article in a plurality of articles for which a scalar score has been derived:
- using the scalar score, in combination with other features, as input to a click-prediction model to derive a ranking score with the click-prediction model, the click-prediction model for ranking content to be presented via a heterogenous feed of the social networking service;
- and
- presenting, by the computer system, the article to the member within the heterogeneous feed of the social networking service in response to the ranking score exceeding a predetermined value, wherein the order of presentation of articles in the heterogeneous feed is determined based on the respective ranking score of each article in the plurality of articles.
6 more elements
A system comprising:
- a processor;
- and
- memory coupled to the processor and storing instructions that, when executed by the processor, cause the system to:
- retrieve, from a database in communication with the system, a member profile of a member of a social networking service, the member profile including information about the member, said information provided by the member;
- identify a plurality of features associated with the member based on the member profile;
- identify a plurality of features associated with an article;
- determine, based on the plurality of features associated with the member and the plurality of features associated with the article, a scalar score reflecting a level of affinity of the member for the article;
- and
- for each article in a plurality of articles for which a scalar score has been derived:
- use the scalar score, in combination with other features, as input to a click-prediction model to derive a ranking score with the click-prediction model, the click-prediction model for ranking content to be presented via a heterogenous feed of the social networking service;
- and
- present the article to the member within a heterogeneous feed of the social networking service in response to the ranking score for the article exceeding a predetermined value, wherein the order of presentation of the plurality of articles in the heterogeneous feed is determined based on the respective ranking score of each article in the plurality of articles.
9 more elements
Dependencies are inferred from the claim text.