Inferring a social connection
A system, method, and apparatus are provided for automatically establishing an inferred (‘follow’) relationship between a first member and a second member of a user community. Based on passive and/or active signals indicating affinity of the first member for the second member, the system determines whether an inferred connection from the first member to the second member would improve the first member's network within the user community. A potential improvement may be observed if the potential connection would improve some aspect or characteristic of a target or ideal network identified for the first member, and/or if there is more than a threshold level of affinity between the members. The first member's network may be pruned, if necessary, to accommodate a connection to the second member if, for example, the connection would violate a constraint associated with the target or ideal network.
1. A method comprising:
receiving via a graphical user interface (GUI) one or more interactions by a first member of a user community with content associated with a second member of the user community;
determining with a machine learning model that an inferred connection from the first member to the second member would improve a first connection network, of the first member, within the user community, wherein the determining comprises:
identifying a member type of the first member;
obtaining a set of characteristics of a target network associated with the identified member type, wherein the target network is deduced by examining a plurality of other members of the identified member type, and wherein the set of characteristics includes a number of connections within the target network; and
determining that addition of the inferred connection to the first connection network will cause corresponding characteristics of the first connection network to more closely match the set of characteristics of the target network;
responsive to determining that addition of the inferred connection would improve the first connection network, modifying the first connection network to include the inferred connection to the second member, without modifying a second connection network of the second member, wherein said modifying comprises:
updating a stored graph of the first connection network to include the second member and the inferred connection; and
causing a content feed to be displayed to the first member that includes content with which the second member interacted, the content with which the second member interacted included in the content feed based on said modifying of the first connection network to include the inferred connection to the second member.
2. The method of claim 1 , wherein identifying a member type of the first member comprises:
examining multiple attributes of a profile of the first member;
wherein the member type is one of:
professional;
professor;
student;
sales;
public face; and
thought leader.
3. The method of claim 1 , wherein the set of characteristics comprises one or more of:
a number of inferred connections within the target network;
a maximum number of connections, within the target network, to members having a predetermined member type; and
a minimum number of connections, within the target network, to members having diversity within a predetermined profile attribute.
4. The method of claim 1 , wherein determining whether an inferred connection from the first member to the second member would improve the first connection network comprises:
for each of the interactions by the first member with content associated with the second member:
assigning a value to the content interaction; and
calculating a product of the assigned weight and a weight corresponding to the interaction; and
combining the calculated products to determine an active affinity of the first member for the second member.
5. The method of claim 4 , wherein determining whether an inferred connection from the first member to the second member would improve the first connection network further comprises:
determining a passive affinity of the first member with the second member based on characteristics of one or more other members of the user community connected to both the first member and the second member;
combining the active affinity and the passive affinity to yield a value; and
when the combined value exceeds a threshold, determining that the inferred connection from the first member to the second member would improve the first connection network.
6. The method of claim 1 , wherein observing one or more interactions of the first member with content associated with the second member comprises:
receiving notification of a first interaction of the first member with a first content item with which the second member previously interacted; and
determining a content category associated with the first interaction.
7. The method of claim 6 , wherein the content category is one of:
a content creation category;
a content curation category;
a content consumption category; and
a profile update category.
8. The method of claim 1 , further comprising:
based on member types of the first member and the second member, characterizing the inferred connection as one of the following types:
colleague;
competitor;
mentor;
poaching; and
business partner.
9. The method of claim 1 , further comprising:
responsive to determining that addition of the inferred connection from the first member to the second member would improve the first connection network, of the first member, within the user community:
determining that addition of the inferred connection to the first connection network will violate a characteristic of the target network; and
removing from the first connection network another inferred connection from the first member to a third member.
10. An apparatus, comprising:
one or more processors; and
memory storing instructions that, when executed by the one or more processors, cause the apparatus to:
receive via a graphical user interface (GUI) one or more interactions by a first member of a user community with content associated with a second member of the user community;
determine with a machine learning model that an inferred connection from the first member to the second member would improve a first connection network, of the first member, within the user community, wherein the determining comprises:
identifying a member type of the first member;
obtaining a set of characteristics of a target network associated with the identified member type, wherein the target network is deduced by examining a plurality of other members of the identified member type, and wherein the set of characteristics includes a number of connections within the target network; and
determining that addition of the inferred connection to the first connection network will cause corresponding characteristics of the first connection network to more closely match the set of characteristics of the target network;
responsive to determining that addition of the inferred connection would improve the first connection network, modify the first connection network to include the inferred connection to the second member, without modifying a second connection network of the second member, wherein said modifying comprises:
updating a stored graph of the first connection network to include the second member and the inferred connection; and
causing a content feed to be displayed to the first member that includes content with which the second member interacted, the content with which the second member interacted included in the content feed based on said modifying of the first connection network to include the inferred connection to the second member.
11. The apparatus of claim 10 , wherein identifying a member type of the first member comprises:
examining multiple attributes of a profile of the first member;
wherein the member type is one of:
professional;
professor;
student;
sales;
public face; and
thought leader.
12. The apparatus of claim 10 , wherein the set of characteristics comprises one or more of:
a number of inferred connections within the target network;
a maximum number of connections, within the target network, to members having a predetermined member type; and
a minimum number of connections, within the target network, to members having diversity within a predetermined profile attribute.
13. The apparatus of claim 10 , wherein determining whether an inferred connection from the first member to the second member would improve the first connection network comprises:
for each of the interactions by the first member with content associated with the second member:
assigning a value to the content interaction; and
calculating a product of the assigned weight and a weight corresponding to the interaction; and
combining the calculated products to determine an active affinity of the first member for the second member.
14. The apparatus of claim 13 , wherein determining whether an inferred connection from the first member to the second member would improve the first connection network further comprises:
determining a passive affinity of the first member with the second member based on characteristics of one or more other members of the user community connected to both the first member and the second member;
combining the active affinity and the passive affinity to yield a value; and
when the combined value exceeds a threshold, determining that the inferred connection from the first member to the second member would improve the first connection network.
15. The apparatus of claim 10 , wherein observing one or more interactions of the first member with content associated with the second member comprises:
receiving notification of a first interaction of the first member with a first content item with which the second member previously interacted; and
determining a content category associated with the first interaction.
16. The apparatus of claim 15 , wherein the content category is one of:
a content creation category;
a content curation category;
a content consumption category; and
a profile update category.
17. The apparatus of claim 10 , wherein the memory further stores instructions that, when executed by the one or more processors, cause the apparatus to:
based on member types of the first member and the second member, characterizing the inferred connection as one of the following types:
colleague;
competitor;
mentor;
poaching; and
business partner.
18. The apparatus of claim 10 , wherein the memory further stores instructions that, when executed by the one or more processors, cause the apparatus to:
responsive to determining that addition of the inferred connection from the first member to the second member would improve the first connection network, of the first member, within the user community:
determine that addition of the inferred connection to the first connection network will violate a characteristic of the target network; and
remove from the first connection network another inferred connection from the first member to a third member.
19. A system, comprising:
one or more processors; and
an inference module comprising a non-transitory computer-readable medium storing instructions that, when executed, cause the system to:
receive via a graphical user interface (GUI) one or more interactions by a first member of a user community with content associated with a second member of the user community; and
determine with a machine learning model that an inferred connection from the first member to the second member would improve a first connection network, of the first member, within the user community, wherein the determining comprises:
identifying a member type of the first member;
obtaining a set of characteristics of a target network associated with the identified member type, wherein the target network is deduced by examining a plurality of other members of the identified member type, and wherein the set of characteristics includes a number of connections within the target network; and
determining that addition of the inferred connection to the first connection network will cause corresponding characteristics of the first connection network to more closely match the set of characteristics of the target network;
a connection module comprising a non-transitory computer-readable medium storing instructions that, when executed, cause the system to:
responsive to determining that addition of the inferred connection would improve the first connection network, modify the first connection network to include the inferred connection to the second member, without modifying a second connection network of the second member, wherein said modifying comprises:
updating a stored graph of the first connection network to include the second member and the inferred connection; and
a content-serving module comprising a non-transitory computer-readable medium storing instructions that, when executed, cause the system to:
causing a content feed to be displayed to the first member that includes content with which the second member interacted, the content with which the second member interacted included in the content feed based on said modifying of the first connection network to include the inferred connection to the second member.
20. The system of claim 19 , wherein the non-transitory computer-readable medium of the inference module further stores instructions that, when executed, cause the system to:
responsive to determining that addition of the inferred connection from the first member to the second member would improve the first connection network, of the first member, within the user community:
determine that addition of the inferred connection to the first connection network will violate a characteristic of the target network; and
remove from the first connection network another inferred connection from the first member to a third member.