Systems and methods to match job candidates and job titles based on machine learning model

1. A computer-implemented method comprising:

receiving, by a computing system, a resume corpus and selected job titles;

training, by the computing system, a machine learning model based on terms from the resume corpus, wherein the model is based at least in part on a technique that creates a set of vector representations of the terms from the resume corpus in a vector space based on semantics of the terms;

converting, by the computing system, based on the model, the selected job titles to a plurality of vector representations in the vector space based on semantics of the job titles, wherein the plurality of vector representations of the selected job titles constitute a plurality of anchor points in the vector space;

organizing, by the computing system, profile information of a user of a social networking system into profile information types that include at least one of profession, educational institution, educational major, or educational degrees;

identifying, by the computing system, based on the model, vector representations from the set of vector representations in the vector space for at least one term of at least one profile information type of a user of a social networking system; and

selecting, by the computing system, a first anchor point representing a first job title among the selected job titles matched to the user,

wherein the selecting is based on the vector representations in the vector space for the at least one term of the at least one profile information type of the user and the plurality of anchor points in the vector space, and

wherein the selecting is further based on a rule that assigns hierarchical importance levels for the profile information types.

2. The computer-implemented method of claim 1 , wherein the user is associated with an employee of an organization recruiting for the job title.

3. The computer-implemented method of claim 2 , wherein the user and the employee are connections on the social networking system.

4. The computer-implemented method of claim 1 , wherein the selecting further comprises:

calculating a pairwise distance between each vector representation of each term of each profile information type and each anchor point.

5. The computer-implemented method of claim 4 , wherein the selecting further comprises:

for each term of each profile information type, identifying an anchor point that is nearest to the term; and

for each profile information type, determining an identified anchor point that is identified most frequently.

6. The computer-implemented method of claim 5 , wherein the identifying an anchor point further comprises:

applying a threshold distance value; and

discarding an anchor point identified from a calculation of pairwise distance when the pairwise distance does not satisfy the threshold distance value.

7. A system comprising:

at least one processor; and

a memory storing instructions that, when executed by the at least one processor, cause the system to perform:

receiving a resume corpus and selected job titles;

training a machine learning model based on terms from the resume corpus, wherein the model is based at least in part on a technique that creates a set of vector representations of the terms from the resume corpus in a vector space based on semantics of the terms;

converting based on the model, the selected job titles to a plurality of vector representations in the vector space based on semantics of the job titles, wherein the plurality of vector representations of the selected job titles constitute a plurality of anchor points in the vector space;

organizing profile information of a user of a social networking system into profile information types that include at least one of profession, educational institution, educational major, or educational degrees;

identifying, based on the model, vector representations from the set of vector representations in the vector space for at least one term of at least one profile information type of a user of a social networking system; and

selecting a first anchor point representing a first job title among the selected job titles matched to the user,

wherein the selecting is based on the vector representations in the vector space for the at least one term of the at least one profile information type of the user and the plurality of anchor points in the vector space, and

wherein the selecting is further based on a rule that assigns hierarchical importance levels for the profile information types.

8. The system of claim 7 , wherein the user is associated with an employee of an organization recruiting for the job title.

9. The system of claim 8 , wherein the user and the employee are connections on the social networking system.

10. The system of claim 7 , wherein the selecting further comprises:

calculating a pairwise distance between each vector representation of each term of each profile information type and each anchor point.

11. The system of claim 10 , wherein the selecting further comprises:

for each term of each profile information type, identifying an anchor point that is nearest to the term; and

for each profile information type, determining an identified anchor point that is identified most frequently.

12. The system of claim 11 , wherein the identifying an anchor point further comprises:

applying a threshold distance value; and

discarding an anchor point identified from a calculation of pairwise distance when the pairwise distance does not satisfy the threshold distance value.

13. A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computing system, cause the computing system to perform a method comprising:

receiving a resume corpus and selected job titles;

training a machine learning model based on terms from the resume corpus, wherein the model is based at least in part on a technique that creates a set of vector representations of the terms from the resume corpus in a vector space based on semantics of the terms;

converting based on the model, the selected job titles to a plurality of vector representations in the vector space based on semantics of the job titles, wherein the plurality of vector representations of the selected job titles constitute a plurality of anchor points in the vector space;

organizing profile information of a user of a social networking system into profile information types that include at least one of profession, educational institution, educational major, or educational degrees;

identifying, based on the model, vector representations from the set of vector representations in the vector space for at least one term of at least one profile information type of a user of a social networking system; and

selecting a first anchor point representing a first job title among the selected job titles matched to the user,

wherein the selecting is based on the vector representations in the vector space for the at least one term of the at least one profile information type of the user and the plurality of anchor points in the vector space, and

wherein the selecting is further based on a rule that assigns hierarchical importance levels for the profile information types.

14. The non-transitory computer-readable storage medium of claim 13 , wherein the user is associated with an employee of an organization recruiting for the job title.

15. The non-transitory computer-readable storage medium of claim 14 , wherein the user and the employee are connections on the social networking system.

16. The non-transitory computer-readable storage medium of claim 13 , wherein the selecting further comprises:

calculating a pairwise distance between each vector representation of each term of each profile information type and each anchor point.

17. The non-transitory computer-readable storage medium of claim 16 , wherein the selecting further comprises:

for each term of each profile information type, identifying an anchor point that is nearest to the term; and

for each profile information type, determining an identified anchor point that is identified most frequently.

18. The non-transitory computer-readable storage medium of claim 17 , wherein the identifying an anchor point further comprises:

applying a threshold distance value; and

discarding an anchor point identified from a calculation of pairwise distance when the pairwise distance does not satisfy the threshold distance value.