Systems and methods for identifying subject-matter experts
The disclosed computer-implemented method for identifying subject-matter experts may include (i) collecting, by the computing device, a plurality of electronic messages transmitted within an organization, (ii) creating a message graph for the organization, (iii) extracting a plurality of topics from the plurality of electronic messages transmitted within the organization, (iv) annotating the message graph by correlating each topic within the plurality of topics with each edge of the message graph that represents an electronic message related to the topic, and (v) identifying, based on an analysis of the annotated message graph, at least one vertex that represents an expert on at least one topic from the plurality of topics. Various other methods, systems, and computer-readable media are also disclosed.
1. A computer-implemented method for identifying subject-matter experts, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising:
collecting, by the computing device, a plurality of electronic messages transmitted within an organization;
creating a message graph for the organization wherein:
each vertex of the message graph represents at least one of a sender of an electronic message within the plurality of electronic messages and a recipient of an electronic message within the plurality of electronic messages; and
each edge of the message graph represents at least one electronic message within the plurality of electronic messages and connects a vertex representing a sender of the at least one electronic message with a vertex representing a recipient of the at least one electronic message;
extracting a plurality of topics from the plurality of electronic messages transmitted within the organization;
annotating the message graph by correlating each topic within the plurality of topics with each edge of the message graph that represents an electronic message related to the topic; and
identifying, based on an analysis of the annotated message graph, at least one vertex that represents an expert on at least one topic from the plurality of topics, wherein identifying the at least one vertex comprises:
specifying at least two topics;
identifying at least one vertex that represents an expert on all of the at least two topics;
weighting the at least two topics as having a different importance to the expert on all of the at least two topics, wherein one of the at least two topics has a higher importance to the expert than another one of the at least two topics;
determining a weighted expertise ranking of experts in the at least two topics; and
identifying the expert based on the one of the at least two topics having the higher importance to the expert and the weighted expertise ranking of the experts in the at least two topics.
2. The computer-implemented method of claim 1 , wherein identifying, based on the analysis of the annotated message graph, the at least one vertex that represents an expert on the at least one topic comprises:
isolating a sub-graph of the annotated message graph that comprises electronic messages related to the at least one topic and does not comprise electronic messages not related to the at least one topic; and
analyzing the sub-graph to identify the at least one vertex that represents an expert on the at least one topic.
3. The computer-implemented method of claim 1 , wherein identifying, based on the analysis of the annotated message graph, the at least one vertex that represents an expert on the at least one topic comprises generating a ranked list of vertices that represent experts on the at least one topic.
4. The computer-implemented method of claim 1 , wherein extracting the plurality of topics from the plurality of electronic messages comprises programmatically determining, by the computing device, the plurality of topics.
5. The computer-implemented method of claim 1 , wherein collecting, by the computing device, the plurality of electronic messages transmitted within the organization comprises collecting a predefined percentage of a total number of electronic messages transmitted within the organization.
6. The computer-implemented method of claim 1 , wherein collecting, by the computing device, the plurality of electronic messages transmitted within the organization comprises periodically collecting new electronic messages transmitted within the organization.
7. The computer-implemented method of claim 1 , wherein:
annotating the message graph comprises labeling each edge of the message graph with a timestamp of at least one electronic message represented by the edge; and
identifying, based on the analysis of the annotated message graph, the at least one vertex that represents an expert on the at least one topic comprises basing the analysis of the annotated message graph at least in part on at least one timestamp of at least one edge of the annotated message graph.
8. The computer-implemented method of claim 1 , further comprising analyzing the annotated message graph in order to identify at least one important topic that is correlated with an amount of edges that meets a predetermined threshold for topic importance.
9. The computer-implemented method of claim 1 , wherein the at least two topics comprise at least two different security topics.
10. A system for identifying subject-matter experts, the system comprising:
a collection module, stored in memory, that collects, by a computing device, a plurality of electronic messages transmitted within an organization;
a creation module, stored in memory, that creates a message graph for the organization wherein:
each vertex of the message graph represents at least one of a sender of an electronic message within the plurality of electronic messages and a recipient of an electronic message within the plurality of electronic messages; and
each edge of the message graph represents at least one electronic message within the plurality of electronic messages and connects a vertex representing a sender of the at least one electronic message with a vertex representing a recipient of the at least one electronic message;
an extraction module, stored in memory, that extracts a plurality of topics from the plurality of electronic messages transmitted within the organization;
an annotating module, stored in memory, that annotates the message graph by correlating each topic within the plurality of topics with each edge of the message graph that represents an electronic message related to the topic;
an identification module, stored in memory, that identifies, based on an analysis of the annotated message graph, at least one vertex that represents an expert on at least one topic from the plurality of topics, wherein the identification module identifies the at least one vertex by:
specifying at least two topics;
identifying at least one vertex that represents an expert on all of the at least two topics;
weighting the at least two topics as having a different importance to the expert on all of the at least two topics, wherein one of the at least two topics has a higher importance to the expert than another one of the at least two topics;
determining a weighted expertise ranking of experts in the at least two topics; and
identifying the expert based on the one of the at least two topics having the higher importance to the expert and the weighted expertise ranking of the experts in the at least two topics; and
at least one physical processor configured to execute the collection module, the creation module, the extraction module, the annotating module, and the identification module.
11. The system of claim 10 , wherein the identification module identifies, based on the analysis of the annotated message graph, the at least one vertex that represents an expert on the at least one topic by:
isolating a sub-graph of the annotated message graph that comprises electronic messages related to the at least one topic and does not comprise electronic messages not related to the at least one topic; and
analyzing the sub-graph to identify the at least one vertex that represents an expert on the at least one topic.
12. The system of claim 10 , wherein the identification module identifies, based on the analysis of the annotated message graph, the at least one vertex that represents an expert on the at least one topic by generating a ranked list of vertices that represent experts on the at least one topic.
13. The system of claim 10 , wherein the extraction module extracts the plurality of topics from the plurality of electronic messages by programmatically determining, by the computing device, the plurality of topics.
14. The system of claim 10 , wherein the collection module collects, by the computing device, the plurality of electronic messages transmitted within the organization by collecting a predefined percentage of a total number of electronic messages transmitted within the organization.
15. The system of claim 10 , wherein the collection module collects, by the computing device, the plurality of electronic messages transmitted within the organization by periodically collecting new electronic messages transmitted within the organization.
16. The system of claim 10 , wherein:
the annotating module annotates the message graph by labeling each edge of the message graph with a timestamp of at least one electronic message represented by the edge; and
the identification module identifies, based on the analysis of the annotated message graph, the at least one vertex that represents an expert on the at least one topic by basing the analysis of the annotated message graph at least in part on at least one timestamp of at least one edge of the annotated message graph.
17. The system of claim 10 , wherein the identification module analyzes the annotated message graph in order to identify at least one important topic that is correlated with an amount of edges that meets a predetermined threshold for topic importance.
18. The system of claim 10 , wherein the at least two topics comprise at least two different security topics.
19. A non-transitory computer-readable medium comprising one or more computer-readable instructions that, when executed by at least one processor of a computing device, cause the computing device to:
collect, by the computing device, a plurality of electronic messages transmitted within an organization;
create a message graph for the organization wherein:
each vertex of the message graph represents at least one of a sender of an electronic message within the plurality of electronic messages and a recipient of an electronic message within the plurality of electronic messages; and
each edge of the message graph represents at least one electronic message within the plurality of electronic messages and connects a vertex representing a sender of the at least one electronic message with a vertex representing a recipient of the at least one electronic message;
extract a plurality of topics from the plurality of electronic messages transmitted within the organization;
annotate the message graph by correlating each topic within the plurality of topics with each edge of the message graph that represents an electronic message related to the topic; and
identify, based on an analysis of the annotated message graph, at least one vertex that represents an expert on at least one topic from the plurality of topics, wherein the at least one vertex is identified by:
specifying at least two topics;
identifying at least one vertex that represents an expert on all of the at least two topics;
weighting the at least two topics as having a different importance to the expert on all of the at least two topics, wherein one of the at least two topics has a higher importance to the expert than another one of the at least two topics;
determining a weighted expertise ranking of experts in the at least two topics; and
identifying the expert based on the one of the at least two topics having the higher importance to the expert and the weighted expertise ranking of the experts in the at least two topics.
20. The non-transitory computer-readable medium of claim 19 , wherein the one or more computer-readable instructions cause the computing device to identify, based on the analysis of the annotated message graph, the at least one vertex that represents an expert on the at least one topic by:
isolating a sub-graph of the annotated message graph that comprises electronic messages related to the at least one topic and does not comprise electronic messages not related to the at least one topic; and
analyzing the sub-graph to identify the at least one vertex that represents an expert on the at least one topic.