Influencer Scoring Model

Abstract

A method for operating an influencer scoring model includes receiving, for a plurality of professionals, professional information and publication information. For each professional, the method also includes generating a professional profile that includes a plurality of attributes. The plurality of attributes are based on (i) professional information and (ii) publication information. The method also includes determining one or more attribute connections among the plurality of attributes of the professional profile. Each attribute connection is based on a common attribute shared between the respective professional and another professional. The method also includes generating a score that includes an authority component and an influence component. A scoring model is configured to receive the plurality of attributes and the attribute connections for the respective professional. The authority component represents at least one attribute corresponding to the received professional information. The influence component represents one or more attribute connections of the professional profile.

What is claimed is: 
     
         1 . A computer-implemented method when executed by data processing hardware causes the data processing hardware to perform operations comprising:
 receiving professional information and publication information for a professional of a plurality of professionals;   generating a professional profile for the professional comprising a plurality of attributes related to the professional, the plurality of attributes derived from (i) the received professional information and (ii) the received publication information;   determining one or more attribute connections among the plurality of attributes of the professional profile, each attribute connection based on a common attribute shared between the professional and one or more other professionals from the plurality of professionals, wherein each attribute connection comprises an influence weight indicating a strength of the common attribute shared between the professional and the one or more other professionals of the plurality of professionals; and   for each respective attribute connection of the one or more attribute connections:
 identifying a time period associated with the respective attribute connection; 
 determining a time difference between the time period associated with the respective attribute connection and a current time; and 
 discounting the influence weight for the respective attribute connection based on the time difference between the time period associated with the respective attribute connection and a current time. 
   
     
     
         2 . The computer-implemented method of  claim 1 , wherein the operations further comprise generating, using a scoring model, a score comprising an authority component and an influence component, the scoring model configured to receive, as inputs, the plurality of attributes related to the professional and the one or more attribute connections for the professional, wherein the authority component representing a first portion of the score determined based on a rank of at least one attribute of the plurality of attributes corresponding to the received professional information of the professional, and wherein the influence component representing a second portion of the score determined based on the strength of the one or more attribute connections among the plurality of attributes of the professional profile of the professional. 
     
     
         3 . The computer-implemented method of  claim 2 , wherein the influence component of the score is generated by:
 for each attribute connection of the one or more attribute connections:
 identifying a duration of the respective attribute connection; and 
 generating the influence weight for the respective attribute connection based on the identified duration of the respective attribute connection, the influence weight further indicating a correlation between the duration of the respective attribute connection and the respective attribute connection; and 
   generating the influence component of the score by aggregating each influence weight associated with the one or more attribute connections.   
     
     
         4 . The computer-implemented method of  claim 2 , wherein the authority component of the score is generated by:
 for each of the at least one attribute of the plurality of attributes corresponding to the received professional information of the professional:   identifying a rank associated with the respective attribute; and   weighing the respective attribute by the identified rank associated with the respective attribute.   
     
     
         5 . The computer-implemented method of  claim 2 , wherein the score comprises a sum of the authority component and the influence component. 
     
     
         6 . The computer-implemented method of  claim 1 , wherein the professional information comprises at least one of a current employer, a past employer, a field of study, or an education degree. 
     
     
         7 . The computer-implemented method of  claim 1 , wherein the publication information corresponds to at least one publication and comprises at least one of:
 an author of the at least one publication;   a coauthor of the at least one publication;   an abstract of the at least one publication;   a title of the at least one publication; or   a mesh term associated with the at least one publication.   
     
     
         8 . The computer-implemented method of  claim 1 , wherein generating the professional profile for the professional comprising the plurality of attributes related to the professional comprises:
 deriving a respective attribute from at least one of the professional information or the publication information;   determining whether the respective attribute corresponds to a preexisting attribute of the professional profile based on the at least one of the professional information or the publication information corresponding to the derived respective attribute; and   when the respective attribute corresponds to the preexisting attribute of the professional profile, updating the professional profile to additionally comprise the respective attribute.   
     
     
         9 . The computer-implemented method of  claim 8 , wherein the respective attribute comprises an alias of the professional profile, and wherein determining whether the respective attribute corresponds to the preexisting attribute of the professional profile comprises determining that the professional information associated with the alias matches the preexisting attribute of the professional profile. 
     
     
         10 . The computer-implemented method of  claim 9 , wherein determining that the professional information associated with the alias matches the preexisting attribute of the professional profile comprises:
 identifying at least one alias of the professional; and   determining, using an alias matching model, whether corresponding professional information associated with the at least one alias matches one of the plurality of attributes of the professional profile, the alias matching model configured to receive, as input, the corresponding professional information for at least one alias of the professional and the plurality of attributes of the professional profile, and to generate, as output, a matching score indicating a likelihood that the at least one alias is an alias of the professional; and   determining that the matching score satisfies a matching score threshold indicating a level of confidence that the at least one alias is an alias of the professional profile.   
     
     
         11 . A system comprising:
 data processing hardware; and   memory hardware in communication with the data processing hardware, the memory hardware storing instructions that when executed on the data processing hardware cause the data processing hardware to perform operations comprising:
 receiving professional information and publication information for a professional of a plurality of professionals; 
 generating a professional profile for the professional comprising a plurality of attributes related to the professional, the plurality of attributes derived from (i) the received professional information and (ii) the received publication information; 
 determining one or more attribute connections among the plurality of attributes of the professional profile, each attribute connection based on a common attribute shared between the professional and one or more other professionals from the plurality of professionals, wherein each attribute connection comprises an influence weight indicating a strength of the common attribute shared between the professional and the one or more other professionals of the plurality of professionals; and 
 for each respective attribute connection of the one or more attribute connections:
 identifying a time period associated with the respective attribute connection; 
 determining a time difference between the time period associated with the respective attribute connection and a current time; and 
 discounting the influence weight for the respective attribute connection based on the time difference between the time period associated with the respective attribute connection and a current time. 
 
   
     
     
         12 . The system of  claim 11 , wherein the operations further comprise generating, using a scoring model, a score comprising an authority component and an influence component, the scoring model configured to receive, as inputs, the plurality of attributes related to the professional and the one or more attribute connections for the professional, wherein the authority component representing a first portion of the score determined based on a rank of at least one attribute of the plurality of attributes corresponding to the received professional information of the professional, and wherein the influence component representing a second portion of the score determined based on the strength of the one or more attribute connections among the plurality of attributes of the professional profile of the professional. 
     
     
         13 . The system of  claim 12 , wherein the influence component of the score is generated by:
 for each attribute connection of the one or more attribute connections:
 identifying a duration of the respective attribute connection; and 
 generating the influence weight for the respective attribute connection based on the identified duration of the respective attribute connection, the influence weight further indicating a correlation between the duration of the respective attribute connection and the respective attribute connection; and 
   generating the influence component of the score by aggregating each influence weight associated with the one or more attribute connections.   
     
     
         14 . The system of  claim 12 , wherein the authority component of the score is generated by:
 for each of the at least one attribute of the plurality of attributes corresponding to the received professional information of the professional:
 identifying a rank associated with the respective attribute; and 
 weighing the respective attribute by the identified rank associated with the respective attribute. 
   
     
     
         15 . The system of  claim 12 , wherein the score comprises a sum of the authority component and the influence component. 
     
     
         16 . The system of  claim 11 , wherein the professional information comprises at least one of a current employer, a past employer, a field of study, or an education degree. 
     
     
         17 . The system of  claim 11 , wherein the publication information corresponds to at least one publication and comprises at least one of:
 an author of the at least one publication;   a coauthor of the at least one publication;   an abstract of the at least one publication;   a title of the at least one publication; or   a mesh term associated with the at least one publication.   
     
     
         18 . The system of  claim 11 , wherein generating the professional profile for the professional comprising the plurality of attributes related to the professional comprises:
 deriving a respective attribute from at least one of the professional information or the publication information;   determining whether the respective attribute corresponds to a preexisting attribute of the professional profile based on the at least one of the professional information or the publication information corresponding to the derived respective attribute; and   when the respective attribute corresponds to the preexisting attribute of the professional profile, updating the professional profile to additionally comprise the respective attribute.   
     
     
         19 . The system of  claim 18 , wherein the respective attribute comprises an alias of the professional profile, and wherein determining whether the respective attribute corresponds to the preexisting attribute of the professional profile comprises determining that the professional information associated with the alias matches the preexisting attribute of the professional profile. 
     
     
         20 . The system of  claim 19 , wherein determining that the professional information associated with the alias matches the preexisting attribute of the professional profile comprises:
 identifying at least one alias of the professional; and   determining, using an alias matching model, whether corresponding professional information associated with the at least one alias matches one of the plurality of attributes of the professional profile, the alias matching model configured to receive, as input, the corresponding professional information for at least one alias of the professional and the plurality of attributes of the professional profile, and to generate, as output, a matching score indicating a likelihood that the at least one alias is an alias of the professional; and   determining that the matching score satisfies a matching score threshold indicating a level of confidence that the at least one alias is an alias of the professional profile.