Methods and systems for identifying member profiles similar to a source member profile

A non-transitory machine-readable storage medium comprising instructions that, when executed by processors of a machine, cause the machine to perform operations comprising:

• receiving a request that specifies a reference profile descriptive of a reference member of a social networking service and that requests identification of candidate profiles similar to the reference profile of the reference member, the candidate profiles each describing a different candidate member of the social networking service;
• determining similarity scores that each correspond to a different candidate profile among the candidate profiles, the determining being performed by one or more processors of the machine and including, for each similarity score being determined:
• accessing a set of portions of a candidate profile among the candidate profiles for comparison to a set of corresponding portions of the reference profile;
• calculating a group of sub-scores that each quantify similarity between a different portion among the set of portions included in the candidate profile and a different corresponding portion in the reference profile;
• and
• determining the similarity score of the candidate profile based on the calculated group of sub-scores;
• and
• causing a presentation of a subset of the candidate profiles in response to the request and based on the subset of the candidate profiles having a predetermined number of highest similarity scores among the candidate profiles.

A system comprising:

• a user interface module comprising one or more processors and configured to receive a request that specifies a reference profile descriptive of a reference member of a social networking service and that requests identification of candidate profiles similar to the reference profile of the reference member, the candidate profiles each describing a different candidate member of the social networking service;
• and
• a recommendation engine comprising one or more processors and configured to determine similarity scores that each correspond to a different candidate profile among the candidate profiles, the determining including, for each similarity score being determined:
• accessing a set of portions of a candidate profile among the candidate profiles for comparison to a set of corresponding portions of the reference profile;
• calculating a group of sub-scores that each quantify similarity between a different portion among the set of portions included in the candidate profile and a different corresponding portion in the reference profile;
• and
• determining the similarity score of the candidate profile based on the calculated group of sub-scores;
• and wherein
• the user interface module comprising one or more processors is configured to cause a presentation of a subset of the candidate profiles in response to the request and based on the subset of the candidate profiles having a predetermined number of highest similarity scores among the candidate profiles.