Social network site recommender system and method
View Patent ↗Particular embodiments evaluate a plurality of messages. Each message may be associated with an author. Information may be logged for each message, such as information associated with the author, information associated with one or more designated recipients of the message, and time information. Correlation values may be determined for one or more sets of designated recipients based on at least a portion of the logged information. Particular embodiments may determine an association amongst a plurality of users over time. The determination may be based on the correlation values. At least one of the plurality of users may comprise at least one of the designated recipients.
1. A method comprising:
evaluating, by one or more processors, a plurality of messages, each message associated with an author;
logging, by the one or more processors, for each message, information associated with the author, information associated with one or more designated recipients of the message, and time information associated with the message;
determining, by the one or more processors, correlation values for one or more sets of the designated recipients based on at least a portion of the logged information; and
determining, by the one or more processors, an association amongst a plurality of users over time, the determining being based on the correlation values, at least one of the plurality of users comprising at least one of the designated recipients.
2. The method of claim 1 , wherein the plurality of users comprises at least one author of an evaluated message.
3. The method of claim 1 , wherein the correlation values are determined based on a desired profile or mathematical curve.
4. The method of claim 3 , wherein the desired profile comprises more recent associations amongst users, and wherein correlation values determined based on more recent time information is weighted more heavily.
5. The method of claim 3 , wherein the desired profile comprises older associations amongst users, and wherein correlation values determined based on older time information is weighted more heavily.
6. The method of claim 3 , wherein the desired profile comprises continuous associations amongst users, and wherein correlation values determined based on a wider range of the time information is weighted more heavily.
7. The method of claim 1 , wherein determining correlation values based on at least a portion of the logged information is limited to messages associated with a selected author or messages associated with other users related to the selected author.
8. The method of claim 1 , further comprising:
analyzing the plurality of messages to classify content therein and form a set of individual content topics; and
processing the plurality of messages to identify a subset of messages related to a first topic;
wherein determining correlation values based on at least a portion of the logged information is limited to logged information associated with messages in the subset of messages.
9. The method of claim 8 , further comprising:
determining, based on the correlation values, a trend over time regarding messages related to a particular topic.
10. The method of claim 9 , further comprising:
determining, based on the correlation values, classifications for a skill set within an organization, wherein the first topic is associated with the skill set, and wherein each message in the subset of messages is associated with the organization.
11. The method of claim 1 , further comprising:
determining, based on the correlation values, one or more trends over time regarding messages associated with one or more users.
12. The method of claim 11 , further comprising:
determining, based on the one or more trends, personnel traits within an organization, wherein determining correlation values based on at least a portion of the logged information is limited to logged information associated with messages associated with the organization.
13. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:
evaluate, by one or more processors, a plurality of messages, each message associated with an author;
log, by the one or more processors, for each message, information associated with the author, information associated with one or more designated recipients of the message, and time information associated with the message;
determine, by the one or more processors, correlation values for one or more sets of the designated recipients based on at least a portion of the logged information; and
determine, by the one or more processors, an association amongst a plurality of users over time, the determining being based on the correlation values, at least one of the plurality of users comprising at least one of the designated recipients.
14. The media of claim 13 , wherein the correlation values are determined based on a desired profile or mathematical curve.
15. The media of claim 14 , wherein the desired profile comprises more recent associations amongst users, and wherein correlation values determined based on more recent time information is weighted more heavily.
16. The media of claim 14 , wherein the desired profile comprises older associations amongst users, and wherein correlation values determined based on older time information is weighted more heavily.
17. The media of claim 14 , wherein the desired profile comprises continuous associations amongst users, and wherein correlation values determined based on a wider range of the time information is weighted more heavily.
18. A system comprising:
one or more processors; and
a memory coupled to the processors comprising instructions executable by the processors, the processors operable when executing the instructions to:
evaluate a plurality of messages, each message associated with an author;
log, for each message, information associated with the author, information associated with one or more designated recipients of the message, and time information associated with the message;
determine correlation values for one or more sets of the designated recipients based on at least a portion of the logged information; and
determine an association amongst a plurality of users over time, the determining being based on the correlation values, at least one of the plurality of users comprising at least one of the designated recipients.
19. The system of claim 18 , the processors being further operable to:
analyze the plurality of messages to classify content therein and form a set of individual content topics; and
process the plurality of messages to identify a subset of messages related to a first topic;
wherein the determination of the correlation values based on at least a portion of the logged information is limited to logged information associated with messages in the subset of messages.
20. The system of claim 18 , the processors being further operable to:
determine, based on the correlation values, a trend over time regarding messages related to a particular topic.