Granted Patent B2

US 12,602,440 · App. 17/950,628 · Granted Apr 14, 2026

Inventors: Xin Fan (Beijing, CN); Liang Wang (Beijing, CN); Hao Zheng (Saratoga, CA)

Assignee: YAHOO ASSETS LLC

G06F16/9535G06F16/24578G06F16/9024H04L67/10H04L67/306

Abstract

Claims (68)

1 . A method, implemented on a computing device having at least one processor, storage, and a communication platform capable of connecting to a network for online user profiling, comprising:

determining, by the computing device, a social media network including a first user and other users in a social graph associated with the first user;

obtaining first web-based content with which the first user interacts;

extracting, from the first web-based content, a first link pointing to second web-based content;

obtaining third web-based content with which the other users interact;

extracting, from the third web-based content, a second link pointing to fourth web-based content;

extracting, by the computing device based on a content analysis platform model, features from the first and second web-based content and features from the third and fourth web-based content, wherein the extracted features comprise publishers of the first, second, third, and fourth web-based content;

scoring each of the publishers based on relevance of the publisher to each of the first, second, third, and fourth web-based content;

generating, by the computing device for the first user, a first user profile based on the scoring;

generating, by the computing device based on user profiles of the other users and weights assigned to the publishers, a baseline user profile indicative of popularity of one or more of the extracted publishers in the social media network;

adjusting, by the computing device, the first user profile based on the baseline user profile;

updating, by the computing device, the first user profile based on a time-decay model, wherein historical weights assigned to the publishers are updated based on the time-decay model, and the time decay model includes a time-decay factor that reduces the historical weights in an exponential manner with respect to a time instance at which the first, second, third, and fourth web-based content are obtained;

generating, based on the updated first user profile, a personalized content recommendation; and

presenting, via an application running on an electronic user device associated with the first user, the personalized content recommendation to the first user.

2 . The method of claim 1 , further comprising:

determining a reference user pool, wherein the reference user pool includes one of the following:

all users of the social media network,

all users in a same group with the first user in the social media network, and

all users connected to and/or followed by the first user in the social media network.

3 . The method of claim 2 , wherein the generating the baseline user profile is based on a user profile of each of the users in the reference user pool.

4 . The method of claim 1 , wherein the baseline user profile indicates background strength of one or more of the extracted features in the social media network.

5 . The method of claim 1 , wherein the features further comprise one or more of entities, categories, and phrases.

6 . A non-transitory, computer-readable medium having information recorded thereon for online user profiling, which, when read by at least one processor, effectuates operations comprising:

determining a social media network including a first user and other users in a social graph associated with the first user;

obtaining first web-based content with which the first user interacts;

extracting, from the first web-based content, a first link pointing to second web-based content;

obtaining third web-based content with which the other users interact;

extracting, from the third web-based content, a second link pointing to fourth web-based content;

extracting, based on a content analysis platform model, features from the first and second web-based content and features from the third and fourth web-based content, wherein the extracted features comprise publishers of the first, second, third, and fourth web-based content;

scoring each of the publishers based on relevance of the publisher to each of the first, second, third, and fourth web-based content;

generating, for the first user, a first user profile based on the scoring;

generating, based on user profiles of the other users and weights assigned to the publishers, a baseline user profile indicative of popularity of one or more of the extracted publishers in the social media network;

adjusting the first user profile based on the baseline user profile;

updating the first user profile based on a time-decay model, wherein historical weights assigned to the publishers are updated based on the time-decay model, and the time decay model includes a time-decay factor that reduces the historical weights in an exponential manner with respect to a time instance at which the first, second, third, and fourth web-based content are obtained;

generating, based on the updated first user profile, a personalized content recommendation; and

presenting, via an application running on an electronic user device associated with the first user, the personalized content recommendation to the first.

7 . The medium of claim 6 , wherein the operations further comprise:

determining a reference user pool, wherein the reference user pool includes one of the following:

all users of the social media network,

all users in a same group with the first user in the social media network, and

all users connected to and/or followed by the first user in the social media network.

8 . The medium of claim 7 , wherein the generating the baseline user profile is based on a user profile of each of the users in the reference user pool.

9 . The medium of claim 6 , wherein the baseline user profile indicates background strength of one or more of the extracted features in the social media network.

10 . The medium of claim 6 , wherein the features further comprise one or more of entities, categories, and phrases.

11 . A system for online user profiling, the system comprising:

memory storing computer program instructions; and

one or more processors that, in response to executing the computer program instructions, effectuate operations comprising:

determining a social media network including a first user and other users in a social graph associated with the first user;

obtaining first web-based content with which the first user interacts;

extracting, from the first web-based content, a first link pointing to second web-based content;

obtaining third web-based content with which the other users interact;

extracting, from the third web-based content, a second link pointing to fourth web-based content;

extracting, based on a content analysis platform model, features from the first and second web-based content and features from the third and fourth web-based content, wherein the extracted features comprise publishers of the first, second, third, and fourth web-based content;

scoring each of the publishers based on relevance of the publisher to each of the first, second, third, and fourth web-based content;

generating, for the first user, a first user profile based on the scoring;

generating, based on user profiles of the other users and weights assigned to the publishers, a baseline user profile indicative of popularity of one or more of the extracted publishers in the social media network;

adjusting the first user profile based on the baseline user profile;

updating the first user profile based on a time-decay model, wherein historical weights assigned to the publishers are updated based on the time-decay model, and the time decay model includes a time-decay factor that reduces the historical weights in an exponential manner with respect to a time instance at which the first, second, third, and fourth web-based content are obtained;

generating, based on the updated first user profile, a personalized content recommendation; and

presenting, via an application running on an electronic user device associated with the first user, the personalized content recommendation to the first.

12 . The system of claim 11 , wherein the operations further comprise:

determining a reference user pool, wherein the reference user pool includes one of the following:

all users of the social media network,

all users in a same group with the first user in the social media network, and

all users connected to and/or followed by the first user in the social media network.

13 . The system of claim 12 , wherein the generating the baseline user profile is based on a user profile of each of the users in the reference user pool.

14 . The system of claim 11 , wherein the baseline user profile indicates background strength of one or more of the extracted features in the social media network.

15 . The system of claim 11 , wherein the features further comprise one or more of entities, categories, and phrases.

Assignments (6)

Continuity (2)

References Cited (21)