Determining a company rank utilizing on-line social network data

A computer-implemented system comprising:

• one or more processors, and
• a non-transitory computer readable storage medium comprising instructions that when executed by the one or processors cause the one or more processors to perform operations comprising:
• accessing a base set of member profiles, a member profile from the base set of member profiles representing a member of an on-line social network system;
• generating a plurality of perturbed sets by repeatedly bootstrap resampling the base set of member profiles;
• a ranking data generator, implemented using at least one processor, to generate ranking data for each company in a set of subject companies, based on the plurality of perturbed sets of member profiles;
• determining respective desirability scores for each company in the set of subject companies based on the ranking data for each company in the set of subject companies;
• and
• storing, in a database, the respective desirability scores for companies in the set of subject companies, wherein the generating of the ranking data for each company in the set of subject companies comprises, for each sample set in the plurality of perturbed sets;
• extracting transition data from a sample set in the plurality of perturbed sets, an item of the transition data indicating transition from a first company to a second company of a member represented by a profile from the given set, the first company and the second company being from the set of subject companies,
• utilizing the transition data for generating a talent flow graph, nodes in the talent flow graph representing respective companies from the set of subject companies, an edge connecting a first node and a second node in the talent flow graph indicating that one or more members represented by respective profiles from the sample set transitioned from a first company represented by the first node to a second company represented by the second node,
• assigning a weight to the edge connecting the first node and the second node, the weight reflecting a number of members represented by respective profiles from the sample set who transitioned from the first company to the second company, and
• applying a ranking algorithm to the talent flow graph to obtain node scores for nodes in the talent flow graph, a node score for a certain node from the nodes in the talent flow graph indicating a likelihood that a member represented by a member profile in the on-line social network system transitions to a company represented by the certain node, the node score for the certain node included in ranking data for the company represented by the certain node.

A machine-readable non-transitory storage medium having instruction data executable by a machine to cause the machine to perform operations comprising:

• accessing a base set of member profiles, a member profile from the base set of member profiles representing a member of an on-line social network system,
• generating a plurality of perturbed sets by repeatedly bootstrap resampling the base set of member profiles;
• based on the plurality of perturbed sets of member profiles generating ranking data for each company in a set of subject companies;
• based on the ranking data for each company in the set of subject companies, determining respective desirability scores for each company in the set of subject companies;
• and
• storing, in a database, the respective desirability scores for companies in the set of subject companies,
• wherein the generating of the ranking data for each company in the set of subject companies comprises, for each sample set in the plurality of perturbed sets;
• extracting transition data from a sample set in the plurality of perturbed sets, an item of the transition data indicating transition from a first company to a second company of a member represented by a profile from the given set, the first company and the second company being from the set of subject companies,
• utilizing the transition data for generating a talent flow graph, nodes in the talent flow graph representing respective companies from the set of subject companies, an edge connecting a first mode and a second node in the talent flow graph indicating that one or more members represented by respective profiles from the same set transitioned from a first company represented by the first node to a second company represented by the second node,
• assigning a weight to the edge connecting the first node and the second node, the weight reflecting a number of members represented by respective profiles from the sample set who transitioned from the first company to the second company, and
• applying a ranking algorithm to the talent flow graph to obtain node scores for nodes in the talent flow graph, a node score for a certain node from the nodes in the talent flow graph indicating a likelihood that a member represented by a member profile in the on-line social network system transitions to a company represented by the certain node, the node score for certain node included in ranking data for the company represented by the certain node.