Methods and systems useful for identifying the most influent social media users in query-based social data streams
First Claim
Patent Images
1. A system comprising:
- a non-transitory machine-readable medium having instructions stored thereon for execution by a processor to perform a method for identifying the most influent social media users in query-based social data streams comprising the steps ofinterfacing with one or more streaming social data sources;
collecting one or more mentions that match a search query;
storing said mentions on a first server;
enhancing said mentions with additional information to produce enhanced mentions, wherein each enhanced mention is assigned an author node;
storing said enhanced mentions on a second server;
filtering said enhanced mentions into filtered streams;
creating a conversation graph from said filtered streams;
calculating influence scores from said conversation graph by iteratively assigning an influence score to each author node of each mention by computing the outreach of one author node towards the rest of the graph based on the individual probabilities found on each author node edge that the next connected node will amplify an action of said author node; and
outputting the most influent social media users.
1 Assignment
0 Petitions
Accused Products
Abstract
The present invention relates to novel methods and implementing systems that afford a user the ability to analyze a certain stream of data, e.g., independent of size, and identify the people that influence the conversation.
12 Citations
20 Claims
-
1. A system comprising:
- a non-transitory machine-readable medium having instructions stored thereon for execution by a processor to perform a method for identifying the most influent social media users in query-based social data streams comprising the steps of
interfacing with one or more streaming social data sources; collecting one or more mentions that match a search query; storing said mentions on a first server; enhancing said mentions with additional information to produce enhanced mentions, wherein each enhanced mention is assigned an author node; storing said enhanced mentions on a second server; filtering said enhanced mentions into filtered streams; creating a conversation graph from said filtered streams; calculating influence scores from said conversation graph by iteratively assigning an influence score to each author node of each mention by computing the outreach of one author node towards the rest of the graph based on the individual probabilities found on each author node edge that the next connected node will amplify an action of said author node; and outputting the most influent social media users. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13)
- a non-transitory machine-readable medium having instructions stored thereon for execution by a processor to perform a method for identifying the most influent social media users in query-based social data streams comprising the steps of
-
14. A method for identifying the most influent social media users in query-based social data streams comprising the steps of
interfacing with one or more streaming social data sources; -
collecting one or more mentions that match a search query; storing said mentions on a first server; enhancing said mentions with additional information to produce enhanced mentions, wherein each enhanced mention is assigned an author node; storing said enhanced mentions on a second server; filtering said enhanced mentions into filtered streams; creating a conversation graph from said filtered streams; calculating influence scores from said conversation graph by iteratively assigning an influence score to each author node of each mention by computing the outreach of one author node towards the rest of the graph based on the individual probabilities found on each author node edge that the next connected node will amplify an action of said author node; and outputting the most influent social media users, such that the most influent social media users for the streaming social data sources are identified. - View Dependent Claims (15, 16, 17, 18, 19, 20)
-
Specification