Back to results
US 8,275,769 B1Granted
System and method for identifying users relevant to a topic of interest
AssigneePalo Alto Research Center Incorporated
InventorsBongwon Suh, Kevin Robert Canini, Peter L. Pirolli, Ed H. Chi
Life of this patent
Estimating the term…
Abstract
A system and method for identifying users relevant to a topic of interest is provided. A query comprising one or more topics is executed against a corpus of messages. Voting users associated with the messages matching the query are identified. A set of candidate users comprising users connected to the voting users is generated. A relevancy score is computed for each candidate user. The candidate users are ranked by their respective relevancy score.
Claims24
A system for identifying users relevant to a topic of interest, comprising:
- a query module to execute a query comprising one or more topics against a corpus of messages and to identify voting users associated with the messages matching the query;
- a candidate generator module to generate a set of candidate users comprising users connected to the voting users;
- a relevancy scorer module to compute a relevancy score for each candidate user in accordance with:
- E[p|fu,Fu]=(fu+α)/(Fu+α+β)
- where E is an estimated probability p, fu, is a number of the voting users connected to the candidate user, Fu is a total number of users comprising the voting users and non-voting users connected to the candidate user, and α and β are binomials;
- a candidate ranking module to rank the candidate users by their respective relevancy score;
- and
- a processor to execute the modules.
5 more elements
A method for identifying users relevant to a topic of interest, comprising:
- executing a query comprising one or more topics against a corpus of messages;
- identifying voting users associated with the messages matching the query;
- generating a set of candidate users comprising users connected to the voting users;
- computing a relevancy score for each candidate user in accordance with:
- E[p|fu,Fu]=(fu+α)/(Fu+α+β)
- where E is an estimated probability p, fu, is a number of the voting users connected to the candidate user, Fu is a total number of users comprising the voting users and non-voting users connected to the candidate user, and a and are binomials;
- and
- ranking the candidate users by their respective relevancy score.
5 more elements
A system for identifying users relevant to a topic of interest, comprising:
- a query module to execute a query comprising one or more topics against a corpus of messages and to identify voting users associated with the messages matching the query;
- a candidate generator module to generate a set of candidate users comprising further users connected to the voting users;
- a relevancy scorer module to determine a relevancy score for each candidate user, comprising:
- a topic module to build a topic model to associate each candidate user and associated message with one or more topics;
- and
- a relevancy calculation module to compute the relevancy score in accordance with:
- Score
- LDA
-
- (
- user
- i
- )
- =
- ∑
- k
-
- P
- 1
-
- (
- queryterm
- |
- topic
- k
- )
- *
- P
- 2
-
- (
- topic
- k
- |
- user
- i
- )
- where P1(query term|topick) is a probability distribution of terms for each topic, k is a number of the topics, P2(topick|useri) is a probability distribution of the topics for each message of one of the users, and i is a number of the users;
- a candidate ranking module to rank the candidate users by their respective relevancy score;
- and
- a processor to execute the modules.
38 more elements
A method for identifying users relevant to a topic of interest, comprising:
- executing a query comprising one or more topics against a corpus of messages and to identify voting users associated with the messages matching the query;
- generating a set of candidate users comprising further users connected to the voting users;
- determining a relevancy score for each candidate user, comprising:
- building a topic model to associate each candidate user and associated message with one or more topics;
- and
- computing the relevancy score in accordance with:
- Score
- LDA
-
- (
- user
- i
- )
- =
- ∑
- k
-
- P
- 1
-
- (
- queryterm
- |
- topic
- k
- )
- *
- P
- 2
-
- (
- topic
- k
- |
- user
- i
- )
- where P1(query term|topick) is a probability distribution of terms for each topic, k is a number of the topics, P2(topick|useri) is a probability distribution of the topics for each message of one of the users, and i is a number of the users;
- and
- ranking the candidate users by their respective relevancy score.
37 more elements
A system for identifying users relevant to a topic of interest, comprising:
- a query module to execute a query comprising one or more topics against a corpus of messages and to identify voting users associated with the messages matching the query;
- a candidate generator module to generate a set of candidate users comprising further users connected to the voting users;
- a relevancy scorer module to compute a relevancy score for each candidate user;
- a candidate ranking module to rank the candidate users by their respective relevancy score;
- a topic module to identify one or more messages of each candidate user and to apply a topic model to the identified messages;
- a re-ranking module to re-rank each candidate user based on the topic model in accordance with:
- ScoreCombined=WLinkStructure*ScoreLinkStructure+WLDA*ScoreLDA,
- where ScoreLinkStructure equals the relevancy score, ScoreLDA equals the topic model, and 0<WLinkStructure, WLDA<1 and WLinkStructure+WLDA=1;
- and
- a processor to execute the modules.
7 more elements
A method for identifying users relevant to a topic of interest, comprising:
- executing a query comprising one or more topics against a corpus of messages and to identify voting users associated with the messages matching the query;
- generating a set of candidate users comprising further users connected to the voting users;
- computing a relevancy score for each candidate user;
- ranking the candidate users by their respective relevancy score;
- identifying one or more messages of each candidate user and to apply a topic model to the identified messages;
- and
- re-ranking each candidate user based on the topic model in accordance with:
- ScoreCombined=WLinkStructure*ScoreLinkStructure+WLDA*ScoreLDA,
- where ScoreLinkstructure equals the relevancy score, ScoreLDA equals the topic model, and 0<WLinkStructure, WLDA<1 and WLinkStructure+WLDA=1.
6 more elements
Dependencies are inferred from the claim text.