Back to results
US 11,455,351 B2Granted
Method and system for online user profiling
AssigneeYahoo Assets LLC
InventorsXin Fan, Liang Wang, Hao Zheng
Life of this patent
Estimating the term…
Abstract
The present teaching relates to online user profiling. In one example, content associated with a first user of a social media network is obtained. From the content associated with the first user, a first link to a first piece of content is identified. A second user of the social media network associated with the first user is determined in the context of the social media network. From content associated with the second user of the social media network, a second link to a second piece of content is identified. The first and second pieces of content are retrieved based on the first and second links, respectively. User profile of the first user is generated based, at least in part, on the first and second pieces of content.
Claims18
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:
- obtaining first content associated with a first user of a social media network;
- extracting, from the first content, a first link pointing to second content;
- determining a second user of the social media network associated with the first user;
- extracting, from third content associated with the second user, a second link pointing to fourth content;
- retrieving the second content and the fourth content by accessing the first link and the second link, respectively, wherein each of the first and third content is obtained from the social media network and each of the second and fourth content is obtained from a content source not included in the social media network, the first content being different from the third content and the second content being different than the fourth content;
- extracting, from the first, second, third, and fourth content, a plurality of features reflecting a corresponding user's interests, wherein the plurality of features comprise a plurality of publishers associated with the first, second, third and fourth content;
- scoring each of the plurality of publishers based on relevance of the feature to each of the first, second, third, and fourth content;
- generating an aggregated score for each of the plurality of publishers by weighting and aggregating scores of the publisher associated with the first, second, third, and fourth content;
- filtering out at least one of the plurality of publishers based on the aggregated scores of the plurality of features;
- generating a user profile of the first user based on remaining of the plurality of publishers;
- and
- updating the user profile based on a time-decay model, wherein historical weights assigned to the scores of the plurality of 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 time instances at which the first, second, third, and fourth content are obtained.
9 more elements
A system including at least one processor, memory coupled to the at least one processor, and a communication platform capable of connecting to a network for online user profiling comprising:
- a first content retrieving unit implemented by the at least one processor and configured to obtain first content associated with a first user of a social media network;
- a social graph analyzing module implemented by the at least one processor and configured to determine a second user of the social media network associated with the first user;
- a link extraction unit implemented by the at least one processor and configured to extract, from the first content, a first link pointing to second content, and extract, from third content associated with the second user, a second link pointing to fourth content;
- a second content retrieving unit implemented by the at least one processor and configured to retrieve the second content and the fourth content by accessing the first link and the second link, respectively, wherein each of the first and third content is obtained from the social media network and each of the second and fourth content is obtained from a content source not included in the social media network, the first content being different from the third content and the second content being different than the fourth content;
- a feature extraction unit implemented by the at least one processor and configured to extract, from the first, second, third, and fourth content, a plurality of features reflecting a corresponding user's interests, wherein the plurality of features comprise a plurality of publishers associated with the first, second, third and fourth content;
- a feature scoring unit implemented by the at least one processor and configured to score each of the plurality of publishers based on relevance of the feature to each of the first, second, third, and fourth content;
- a feature aggregation unit implemented by the at least one processor and configured to:
- generate an aggregated score for each of the plurality of publishers by aggregating scores of the feature associated with the first, second, third, and fourth content;
- and
- filter out at least one of the plurality of publishers based on the aggregated scores of the plurality of publishers;
- and
- a user interest scoring unit implemented by the at least one processor and configured to:
- generate a user profile of the first user based on remaining of the plurality of publishers;
- and
- update the user profile based on a time-decay model, wherein historical weights assigned to the scores of the plurality of 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 time instances at which the first, second, third, and fourth content are obtained.
12 more elements
A non-transitory, machine-readable medium having information recorded thereon for online user profiling, wherein the information, when read by a machine, causes the machine to perform the following:
- obtaining first content associated with a first user of a social media network;
- extracting, from the first content, a first link pointing to second content;
- determining a second user of the social media network associated with the first user;
- extracting, from third content associated with the second user, a second link pointing to fourth content;
- retrieving the second content and the fourth content by accessing the first link and the second link, respectively, wherein each of the first and third content is obtained from the social media network and each of the second and fourth content is obtained from a content source not included in the social media network, the first content being different from the third content and the second content being different than the fourth content;
- extracting, from the first, second, third, and fourth content, a plurality of features reflecting a corresponding user's interests, wherein the plurality of features comprise a plurality of publishers associated with the first, second, third and fourth content;
- scoring each of the plurality of publishers based on relevance of the feature to each of the first, second, third, and fourth content;
- generating an aggregated score for each of the plurality of publishers by aggregating scores of the feature associated with the first, second, third, and fourth content;
- filtering out at least one of the plurality of publishers based on the aggregated scores of the plurality of publishers;
- generating a user profile of the first user based on remaining of the plurality of publishers;
- and
- updating the user profile based on a time-decay model, wherein historical weights assigned to the scores of the plurality of 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 time instances at which the first, second, third, and fourth content are obtained.
9 more elements
Dependencies are inferred from the claim text.