Computerized system and method for determining non-redundant tags from a user's network activity
Descriptive data relating to at least a subset of a plurality of entities on a website is retrieved over a network. Endorsement data relating to the plurality of entities is retrieved from the website. A first set of probabilities is determined reflecting a probability that endorsements can be attributed to specific aspects. A second set of probabilities is determined reflecting a probability that terms can be attributed to aspects. Using the first set of probabilities and the second set of probabilities, a subset of the terms that are most probably associated with each entity are selected. Tags are then generated for each entity using the selected terms.
1. A method comprising:
identifying, via a computing device, user generated content (UGC) associated with a first user social networking profile comprising digital content provided to a set of other users via a social network, said set of other users comprising profiles that follow the first user profile on the social network;
analyzing, via the device, the UGC and determining topics of the first user profile based on the UGC;
computing, via the device, a term-based representation of each topic, wherein the term-based representation is a measure of probability of a term being associated the topic;
extracting, via the device, a tag for each term-based representation, each tag representing a facet of the first user profile that led each of the other user profiles to follow the first user profile;
identifying, via the device, digital content items based on each tag;
communicating, via the device, said digital content items to each of the other user profiles such that each of the other user profiles receives a digital content item associated with the facet that led the other user profiles to follow the first user profile;
determining a social endorsement graph of the UGC, said graph comprising a visualization of messages sent to and from the first user profile and each other user profiles, the tags and terms related to the term-based representations; and
determining an aspect-entity distribution among the graph by mining the graph to identify which other user profiles are associated with which messages visualized in the graph.
2. The method of claim 1 , wherein said extraction of the tags is based on said determined aspect-entity distribution, wherein said facets associated with each tag correspond to aspects of the UGC.
3. The method of claim 2 , wherein at least one aspect comprises text within the UGC that references at least one topic.
4. The method of claim 1 , further comprising: determining a document-specific distribution of text within each message of the UGC of the first user profile, wherein said determination of said topics is based on said distribution.
5. The method of claim 1 , wherein a first portion of said set of other user profiles receives a digital content item associated with a first topic, and a second portion of said set of other user profiles receives a digital content item associated with a second topic, wherein said first topic and said second topic are associated with different facets of activity on said first user social networking profile.
6. The method of claim 1 , wherein said digital content items are provided by a third party content provider.
7. The method of claim 1 , wherein said analysis of the UGC and computation of the term-based representations is based on the device executing Latent Dirichlet Allocation (LDA) software.
8. The method of claim 1 , wherein said UGC is identified across a plurality of social networks.
9. A computing device comprising:
a processor;
a storage medium for tangibly storing thereon program logic for execution by the processor, the program logic comprising:
logic executed by the processor for identifying, via the device, user generated content (UGC) associated with a first user social networking profile comprising digital content provided to a set of other users via a social network, said set of other users comprising profiles that follow the first user profile on the social network;
logic executed by the processor for analyzing, via the device, the UGC and determining topics of the first user profile based on the UGC;
logic executed by the processor for computing, via the device, a term-based representation of each topic, wherein the term-based representation is a measure of probability of a term being associated the topic;
logic executed by the processor for extracting, via the device, a tag for each term-based representation, each tag representing a facet of the first user profile that led each of the other user profiles to follow the first user profile;
logic executed by the processor for identifying, via the device, digital content items based on each tag;
logic executed by the processor for communicating, via the device, said digital content items to each of the other user profiles such that each of the other user profiles receives a digital content item associated with the facet that led the other user profiles to follow the first user profile;
logic executed by the processor for determining a social endorsement graph of the UGC, said graph comprising a visualization of messages sent to and from the first user profile and each of the other user profiles, tags and terms related to the term-based representations; and
logic executed by the processor for determining an aspect-entity distribution among the graph by mining the graph to identify which other user profiles are associated with which messages visualized in the graph, wherein said extraction of the tags is based on said determined aspect-entity distribution, wherein said facets associated with each tag correspond to aspects of the UGC.
10. The computing device of claim 9 , further comprising:
logic executed by the processor for determining a document-specific distribution of text within each message of the UGC of the first user profile, said wherein said determination of said topics is based on said distribution.
11. The computing device of claim 9 , wherein a first portion of said set of other user profiles receives a digital content item associated with a first topic, and a second portion of said set of other user profiles receives a digital content item associated with a second topic, wherein said first topic and said second topic are associated with different facets of activity on said first user social networking profile.
12. The computing device of claim 9 , wherein said digital content items are provided by a third party content provider.
13. The computing device of claim 9 , wherein said analysis of the UGC and computation of the term-based representations is based on the device executing Latent Dirichlet Allocation (LDA) software.
14. The computing device of claim 9 , wherein said UGC is identified across a plurality of social networks.
15. A non-transitory computer-readable storage medium for tangibly storing thereon computer readable instructions, that when executed by a processor of a computing device, perform a method comprising:
identifying, via the device, user generated content (UGC) associated with a first user social networking profile comprising digital content provided to a set of other users via a social network, said set of other users comprising profiles that follow the first user profile on the social network;
analyzing, via the device, the UGC and determining topics of the first user profile based on the UGC;
computing, via the device, a term-based representation of each topic, wherein the term-based representation is a measure of probability of a term being associated the topic;
extracting, via the device, a tag for each term-based representation, each tag representing a facet of the first user profile that led each of the other user profiles to follow the first user profile;
identifying, via the device, digital content items based on each tag;
communicating, via the device, said digital content items to each of the other user profiles such that each of the other user profiles receives a digital content item associated with the facet that led the other user profiles to follow the first user profile;
determining a social endorsement graph of the UGC, said graph comprising a visualization of messages sent to and from the first user and each other user, the tags and terms related to the term-based representations; and
determining an aspect-entity distribution among the graph by mining the graph to identify which other user profiles are associated with which messages visualized in the graph, wherein said extraction of the tags is based on said determined aspect-entity distribution, wherein said facets associated with each tag correspond to aspects of the UGC.
16. The non-transitory computer-readable storage medium of claim 15 , wherein a first portion of said set of other user profiles receives a digital content item associated with a first topic, and a second portion of said set of other user profiles receives a digital content item associated with a second topic, wherein said first topic and said second topic are associated with different facets of activity on said first user social networking profile.
17. The non-transitory computer-readable storage medium of claim 15 , wherein said analysis of the UGC and computation of the term-based representations is based on the device executing Latent Dirichlet Allocation (LDA) software.