Back to results
US 10,474,963 B1Granted
Network-based organization classification
AssigneeHRL Laboratories, LLC
InventorsPatrick Park, Ryan F. Compton, Tsai-Ching Lu
Life of this patent
Estimating the term…
Abstract
Described is a system for discriminating between individual accounts and organization accounts in a social media network. The system processes data from a social media network comprising a set of user accounts. A directed mention network, a bi-directed mention network, and a set of computer-observable network metrics are generated. A classification model is trained using the set of computer-observable network metrics and a set of ground-truth user accounts. The trained classification model is used to classify each user account in the set of user accounts as an individual account or an organization account, resulting in a set of classified user accounts.
Claims12
A system for network-based organization classification, the system comprising:
- one or more processors and a non-transitory memory having instructions encoded thereon such that when the instructions are executed, the one or more processors perform operations of:
- processing data from a social media network comprising a set of user accounts;
- generating a directed mention network;
- generating a set of directed network metrics from the directed mention network;
- extracting a bi-directed mention network from the directed mention network, wherein the bi-directed mention network comprises only pairs of user accounts from the directed mention network that sent a mention record to each other at least once;
- generating a set of bi-directed network metrics from the bi-directed mention network,
- where in generating the set of bi-directed network metrics, out-mention and in-mention frequencies within the bi-directed network are derived, wherein an out-mention is a mention record directed to another user account, and an in-mention is a mention record directed from another user account;
- determining a directed mention network mention to bi-directed network mention frequency ratio;
- training a classification model using the set of directed network metrics, the set of bi-directed network metrics, and a set of ground-truth user accounts;
- classifying, using the trained classification model, each user account in the set of user accounts as an individual account or an organization account, resulting in a set of classified user accounts;
- and
- generating and pushing targeted advertisements over the social media network to classified individual user accounts.
9 more elements
A computer-implemented method for network-based organization classification, comprising an act of:
- causing one or more processors to execute instructions stored on a non-transitory memory such that upon execution, the one or more processors perform operations of:
- processing data from a social media network comprising a set of user accounts;
- generating a directed mention network;
- generating a set of directed network metrics from the directed mention network;
- extracting a bi-directed mention network from the directed mention network, wherein the bi-directed mention network comprises only pairs of user accounts from the directed mention network that sent a mention record to each other at least once;
- generating a set of bi-directed network metrics from the bi-directed mention network,
- where in generating the set of bi-directed network metrics, out-mention and in-mention frequencies within the bi-directed network are derived, wherein an out-mention is a mention record directed to another user account, and an in-mention is a mention record directed from another user account;
- determining a directed mention network mention to bi-directed network mention frequency ratio;
- training a classification model using the set of directed network metrics, the set of bi-directed network metrics, and a set of ground-truth user accounts;
- classifying, using the trained classification model, each user account in the set of user accounts as an individual account or an organization account, resulting in a set of classified user accounts;
- and
- generating and pushing targeted advertisements over the social media network to classified individual user accounts.
9 more elements
A computer program product for network-based organization classification, the computer program product comprising:
- computer-readable instructions stored on a non-transitory computer-readable medium that are executable by a computer having one or more processors for causing the processor to perform operations of:
- processing data from a social media network comprising a set of user accounts;
- generating a directed mention network;
- generating a set of directed network metrics from the directed mention network;
- extracting a bi-directed mention network from the directed mention network, wherein the bi-directed mention network comprises only pairs of user accounts from the directed mention network that sent a mention record to each other at least once;
- generating a set of bi-directed network metrics from the bi-directed mention network,
- where in generating the set of bi-directed network metrics, out-mention and in-mention frequencies within the bi-directed network are derived, wherein an out-mention is a mention record directed to another user account, and an in-mention is a mention record directed from another user account;
- determining a directed mention network mention to bi-directed network mention frequency ratio;
- training a classification model using the set of directed network metrics, the set of bi-directed network metrics, and a set of ground-truth user accounts;
- classifying, using the trained classification model, each user account in the set of user accounts as an individual account or an organization account, resulting in a set of classified user accounts;
- and
- generating and pushing targeted advertisements over the social media network to classified individual user accounts.
9 more elements
Dependencies are inferred from the claim text.