Back to results
US 9,430,755 B2Granted
System and method to enable communication group identification
AssigneeYahoo! Inc.
InventorsJake Hofman, Winter Mason, Duncan James Watts
Life of this patent
Estimating the term…
Abstract
Methods, system and computer readable medium for discovering social groups include extracting activity related data associated with a user's social interactions from a source. The activity related data identifies information related to social contacts used during the social interactions at the source. The activity related data of the user is analyzed to identify co-occurrences of the social contacts. The co-occurrences determine a set of related contacts. The set of related contacts identified from the activity related data are clustered into distinct social groups. The social groups are used by a user to facilitate subsequent interactions.
Claims16
A computer-implemented method for discovering social groups, comprising:
- extracting activity related data associated with a user's social interactions from a source, by a processor, the activity related data identifying information related to social contacts used in the social interactions;
- analyzing the activity related data of the user, by the processor, to identify co-occurrences of the social contacts in the social interactions, wherein the co-occurrences determine sets of related contacts, the analysis further determines relative affiliation strength of social link for each of the social contacts within each set of related contacts;
- and
- clustering the sets of related contacts identified from the activity related data to generate distinct social groups, by the processor, wherein a sequence of social contacts in each of the generated social groups are in accordance to the affiliation strength of each social contact within the respective social groups, the social groups used to facilitate subsequent social interactions,
- wherein when a particular social contact appears in more than one social group, clustering includes associating a tag derived from context of the activity related data to distinctly identify each of the social groups in which the particular social contact appears, each social contact in the respective social groups associated with an indicator that is indicative of affiliation strength within the social group, the indicator providing a visual cue of the affiliation strengths associated with the social contacts within the social group.
2 more elements
A non-transitory computer program product comprising program instructions embodied on a computer readable medium, which when executed by a processor of a computer implements a method for discovering social groups, the program instructions comprising:
- program instructions for extracting activity related data associated with a user's social interactions from a source, the activity related data identifying information related to social contacts used in the social interactions;
- program instructions for analyzing the activity related data of the user to identify co-occurrences of the social contacts in the social interactions, wherein the co-occurrences determine sets of related contacts, the analysis further determines relative affiliation strength of each of the social contacts within each set of related contacts;
- and
- program instructions for clustering the sets of related contacts identified from the activity related data to generate distinct social groups, wherein a sequence of social contacts in each of the generated social groups are in accordance to the affiliation strength of each social contact within the respective social groups, the social groups used to facilitate subsequent social interactions,
- wherein when a particular social contact appears in more than one social group, program instructions for clustering includes program instructions for associating a tag derived from context of the activity related data to distinctly identify each of the social groups in which the particular social contact appears, and program instructions for associating an indicator to each social contact in the respective social groups, the indicator is indicative of affiliation strength within the social group, the indicator used to provide a visual cue of the affiliation strengths of the social contacts within the social group.
2 more elements
A system for discovering social groups, comprising:
- a server executing an algorithm on a processor, wherein the algorithm is configured to,
- extract activity related data associated with a user's social interactions from a source, the activity related data identifying information related to social contacts used in the social interactions;
- analyze the activity related data of the user to identify co-occurrences of the social contacts in the social interactions, wherein the co-occurrences determine sets of related contacts, the analysis further used to determine relative affiliation strength of each of the social contacts within each set of related contacts;
- and
- cluster the set of related contacts identified from the activity related data to generate distinct social groups, the social groups used to facilitate subsequent interactions, wherein a sequence of social contacts in each of the generated social groups are in accordance to the affiliation strength of each social contact within the respective social groups,
- wherein when a particular social contact appears in more than one social group, associate a tag derived from context of the activity related data to distinctly identify each of the social groups in which the particular social contact appears, each social contact in each of the social groups is associated with an indicator that is indicative of affiliation strength within the social group, the indicator is used to provide visual cue of the affiliation strengths of the social contacts within the social group.
3 more elements
Dependencies are inferred from the claim text.