IDENTIFICATION OF AN AUTHORITY-TYPE LEADER IN A SOCIAL NETWORK
Techniques for identification of an authority-type leader in a social network based on reputation are described. According to various embodiments, it is determined that one or more interactions with a first actor of a plurality of actors of an online social network correspond to decisions by other actors of the plurality of actors of the online social network that the first actor is an information authority on a particular topic. Thereafter, a reputation leadership score associated with the first actor and the particular topic is calculated, the reputation leadership score indicating a strength of the first actor's reputation as an information authority on the particular topic among the plurality of actors of the online social network service. The first actor is then classified as an information authority on the particular topic among the plurality of actors of the online social network service, based on the calculated reputation leadership score.
1 . A method comprising:
accessing profile data of a plurality of actors of an online social network service and interaction data describing interactions between the plurality of actors of the online social network service;
determining, based on the interaction data, that one or more interactions with a first actor of the plurality of actors correspond to decisions by other actors of the plurality of actors that the first actor is an information authority on a particular topic;
calculating, by a machine including a memory and at least one processor and based on the profile data and the interaction data, a reputation leadership score associated with the first actor and the particular topic, the reputation leadership score indicating a strength of the first actor's reputation as an information authority on the particular topic among the plurality of actors of the online social network service; and
classifying the first actor as an information authority on the particular topic among the plurality of actors of the online social network service, based on the calculated reputation leadership score.
2 . The method of claim 1 , wherein the calculating the reputation leadership score further comprises:
calculating a plurality of net decision scores for each specific actor in the plurality of actors, each of the net decision scores indicating a strength of each specific actor's decision that the first actor is the information authority on the particular topic; and
generating the reputation leadership score, based on the plurality of net decision scores.
3 . The method of claim 2 , wherein the calculating of the net decision score for each specific actor further comprises calculating a net personal decision score for the specific actor by:
calculating one or more distinct personal decision scores for the specific actor, the personal decision scores corresponding to distinct personal decisions by the specific actor that the first actor is the information authority on the particular topic; and
generating the net personal decision score for the specific actor, based on the one or more distinct personal decision scores.
4 . The method of claim 3 , wherein the calculating of each distinct personal decision score for the specific actor further comprises:
determining the specific actor's reputation as an information authority on the particular topic;
determining a strength of a relationship between the specific actor and the first actor;
determining a decision type of the corresponding personal decision of the specific actor, the decision type corresponding to a type of an interaction between the specific actor and the first actor; and
calculating the distinct personal decision score for the specific actor, based on the specific actor's reputation, the strength of the relationship, and the decision type.
5 . The method of claim 4 , wherein the strength of the relationship is determined based on the duration of the relationship and the type of the relationship, the type of the relationship corresponding to at least one of:
a network connection;
a homophily;
a formal role; and
an informal role.
6 . The method of claim 4 , wherein the decision type corresponds to:
the specific actor viewing, clicking, liking, sharing, following, or commenting on a content item posted by the first actor that is associated with the particular topic.
7 . The method of claim 4 , wherein the decision type corresponds to:
the specific actor endorsing or recommending the first actor for a skill that is associated with the particular topic.
8 . The method of claim 4 , wherein the decision type corresponds to:
the specific actor endorsing or certifying the first actor as having completed an education program associated with the particular topic.
9 . The method of claim 4 , wherein the decision type corresponds to:
the specific actor publishing a content item authored by the first actor that is associated with the particular topic.
10 . The method of claim 4 , wherein the decision type corresponds to:
the specific actor and the first actor co-authoring a publication associated with the particular topic.
11 . The method of claim 4 , wherein the decision type corresponds to:
the specific actor citing a content item authored by the first actor in a publication associated with the particular topic.
12 . The method of claim 4 , wherein the decision type corresponds to:
the specific actor referencing the first actor.
13 . The method of claim 4 , wherein the decision type corresponds to:
the specific actor following the first actor.
14 . The method of claim 3 , wherein the calculating of the net decision score for each specific actor further comprises calculating a net social decision score for the specific actor, the net social decision score representing an evaluation by the specific actor of a third party actor's decision that the first actor is the information authority on the particular topic.
15 . The method of claim 14 , wherein the calculating of the net social decision score for the specific actor further comprises:
calculating a plurality of net decision scores associated with a plurality of third party actors, each of the net decision scores indicating a strength of each third party actor's decision that the first actor is the information authority on the particular topic;
determining a strength of a relationship between each of the third party actors and the specific actor; and
calculating the net social decision score for the specific actor, based on the plurality of net decision scores and the strengths of the relationships.
16 . The method of claim 1 , wherein the first actor is least one of a:
a member of the online social network service; and
an entity of the online social network service, wherein the entity corresponds to at least one of a company entity, a group entity, and an educational entity associated with the online social network service.
17 . The method of claim 1 , further comprising:
accessing an electronic sociograph depicting the plurality of actors of the online social network service as nodes in the sociograph, the sociograph including edges extending between the nodes that represent relationships between the respective actors; and
annotating a node in the sociograph corresponding to the first actor with information indicating that the first actor is classified as the information authority on the particular topic.
18 . The method of claim 1 , further comprising:
determining that the first actor has posted a content item related to the particular topic on the online social network service, the content item corresponding to an article or a comment; and
promoting the content item in a content feed or a list of comments associated with the online social network service.
19 . A system comprising:
a machine including a memory and at least one processor;
a interaction module configured to:
access profile data of a plurality of actors of an online social network service and interaction data describing interactions between the plurality of actors of the online social network service; and
determine, based on the interaction data, that one or more interactions with a first actor of the plurality of actors correspond to decisions by other actors of the plurality of actors that the first actor is an information authority on a particular topic; and
a leadership module, executable by the machine, configured to:
calculate, based on the profile data and the interaction data, a reputation leadership score associated with the first actor and the particular topic, the reputation leadership score indicating a strength of the first actor's reputation as an information authority on the particular topic among the plurality of actors of the online social network service; and
classify the first actor as an information authority on the particular topic among the plurality of actors of the online social network service, based on the calculated reputation leadership score.
20 . A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:
accessing profile data of a plurality of actors of an online social network service and interaction data describing interactions between the plurality of actors of the online social network service;
determining, based on the interaction data, that one or more interactions with a first actor of the plurality of actors correspond to decisions by other actors that the first actor of the plurality of actors is an information authority on a particular topic;
calculating, based on the profile data and the interaction data, a reputation leadership score associated with the first actor and the particular topic, the reputation leadership score indicating a strength of the first actor's reputation as an information authority on the particular topic among the plurality of actors of the online social network service; and
classifying the first actor as an information authority on the particular topic among the plurality of actors of the online social network service, based on the calculated reputation leadership score.