Feedback-based recommendation of member attributes in social networks

1. A method, comprising:

updating a taxonomy of member attributes, the updating comprising obtaining acceptance rates for recommendations of member attributes as profile edits to a set of members in the social network and ranking the member attributes based on the acceptance rates;

obtaining a set of member features associated with a member of a social network and a set of attribute features associated with a set of member attributes, the attribute features including member responses to previous recommendations of member attributes;

analyzing, by one or more computer systems, the member features and the attribute features to predict a propensity of the member to accept recommendations of the member attributes from the taxonomy as profile edits to a member profile of the member, the analyzing comprising:

applying a statistical model to the member features and the attribute features, the statistical model configured for computing a score for each of a plurality of member attributes that represents a predicted propensity of specified members of the social network to accept recommendations of that member attribute;

obtaining a set of scores representing the predicted propensity of the member to accept recommendations of the member attributes as output from the statistical model; and

obtaining a subset of the member attributes associated with a subset of scores that each exceed a threshold, the subset of member attributes being ordered based on the scores:

outputting, based at least in part on the set of scores, a subset of the member attributes as recommended profile edits to the member; and

tracking responses to the recommended profile edits by the member and, based on the responses, updating recommended profile edits for the member.

2. The method of claim 1 , further comprising:

ordering the subset of the member attributes by the scores; and

using the ordering to output the recommended profile edits to the member.

3. The method of claim 1 , wherein outputting, based at least in part on the set of scores, the subset of the member attributes as recommended profile edits to the member further comprises:

selecting a member attribute for outputting in the recommended profile based on an attribute type of the member attribute.

4. The method of claim 1 , wherein the set of member features comprises at least one of:

a current member attribute of the member;

a previous member attribute of the member;

a job-seeking status of the member;

an activity level of the member; and

a member attribute of a connection of the member.

5. The method of claim 1 , wherein the set of attribute features comprises at least one of:

an trending indicator for a member attribute; and

a member response to a recommendation of the member attribute.

6. The method of claim 1 , wherein the set of member features comprises a member segment of the member.

7. The method of claim 1 , wherein the set of member attributes comprises at least one of:

a skill;

a title;

an industry;

a company;

a school;

a publication;

a certification; and

a summary.

8. An apparatus, comprising:

one or more processors; and

memory storing instructions that, when executed by the one or more processors, cause the apparatus to:

update a taxonomy of member attributes, the updating comprising obtaining acceptance rates for recommendations of member attributes as profile edits to a set of members in the social network and ranking the member attributes based on the acceptance rates;

obtain a set of member features associated with a member of a social network and a set of attribute features associated with a set of member attributes, the attribute features including member responses to previous recommendations of member attributes;

analyze the member features and the attribute features to predict a propensity of the member to accept recommendations of the member attributes from the taxonomy as profile edits to a member profile of the member, the analyzing comprising:

applying a statistical model to the member features and the attribute features, the statistical model configured for computing a score for each of a plurality of member attributes that represents a predicted propensity of specified members of the social network to accept recommendations of that member attribute;

obtaining a set of scores representing the predicted propensity of the member to accept recommendations of the member attributes as output from the statistical model; and

obtaining a subset of the member attributes associated with a subset of scores that each exceed a threshold, the subset of member attributes being ordered based on the scores;

output, based at least in part on the set of scores, a subset of the member attributes as recommended profile edits to the member; and

track responses to the recommended profile edits by the member and, based on the responses, updating recommended profile edits for the member.

9. The apparatus of claim 8 , further comprising:

ordering the subset of the member attributes by the scores; and

using the ordering to output the recommended profile edits to the member.

10. The apparatus of claim 8 , wherein outputting, based at least in part on the set of scores, the subset of the member attributes as recommended profile edits to the member further comprises:

selecting a member attribute for outputting in the recommended profile based on an attribute type of the member attribute.

11. The apparatus of claim 8 , wherein the set of member features comprises at least one of:

a current member attribute of the member;

a previous member attribute of the member;

a job-seeking status of the member;

an activity level of the member;

a member attribute of a connection of the member; and

a member segment of the member.

12. The apparatus of claim 8 , wherein the set of attribute features comprises at least one of:

an trending indicator for a member attribute; and

a member response to a recommendation of the member attribute.

13. The apparatus of claim 8 , wherein the set of member attributes comprises at least one of:

a skill;

a title;

an industry;

a company;

a school;

a publication;

a certification; and

a summary.

14. A system, comprising:

an analysis module comprising a non-transitory computer-readable medium comprising instructions that, when executed, cause the system to:

update a taxonomy of member attributes, the updating comprising obtaining acceptance rates for recommendations of member attributes as profile edits to a set of members in the social network and ranking the member attributes based on the acceptance rates;

obtain a set of member features associated with a member of a social network and a set of attribute features associated with a set of member attributes, the attribute features including member responses to previous recommendations of member attributes; and

analyze the member features and the attribute features to predict a propensity of the member to accept recommendations of the member attributes from the taxonomy as profile edits to a member profile of the member, the analyzing comprising:

applying a statistical model to the member features and the attribute features, the statistical model configured for computing a score for each of a plurality of member attributes that represents a predicted propensity of specified members of the social network to accept recommendations of that member attribute;

obtaining a set of scores representing the predicted propensity of the member to accept recommendations of the member attributes as output from the statistical model; and

obtaining a subset of the member attributes associated with a subset of scores that each exceed a threshold, the subset of member attributes being ordered based on the scores;

a management module comprising a non-transitory computer-readable medium comprising instructions that, when executed, cause the system to:

use the predicted propensity to output a subset of the member attributes as recommended profile edits to the member; and

track responses to the recommended profile edits by the member and, based on the responses, updating recommended profile edits for the member.