Systems and methods to rank job candidates based on machine learning model

Patent metadata

Field	Value
Publication number	US-2017193394-A1
Application number	US-201614987648-A
Country	US
Kind code	A1
Filing date	Jan 4, 2016
Priority date	Jan 4, 2016
Publication date	Jul 6, 2017
Grant date	—

Abstract

Systems, methods, and non-transitory computer readable media are configured to determine a training set to train a machine learning model. A feature set for the model is determined. The model is trained based on the training set and the feature set to determine a score reflecting a probability that each user in an evaluation set of users is qualified for employment with an organization. A ranking of users in the evaluation set is provided based on the score determined for each user.

First claim

What is claimed is:

1 . A computer-implemented method comprising:
determining, by a computing system, a training set to train a machine learning model; determining, by the computing system, a feature set for the model; training, by the computing system, the model based on the training set and the feature set to determine a score reflecting a probability that each user in an evaluation set of users is qualified for employment with an organization; and providing, by the computing system, a ranking of users in the evaluation set based on the score determined for each user.

2 . The computer-implemented method of claim 1 , wherein the training set includes users who are employees of the organization.

3 . The computer-implemented method of claim 2 , wherein the training set further includes users who are employees of a second organization of the same organization type as the organization.

4 . The computer-implemented method of claim 1 , wherein the feature set comprises at least one of a number of connections of a user in the training set on a social networking system, a number of days since the user performed an action on the social networking system, a number of requests by the user to initiate connections on the social networking system, a number of entities who are following the user on the social networking system, a number of entities followed by the user on the social networking system, college attended by the user, graduate school attended by the user, degrees obtained by the user, concentrations of study by the user, and employers of the user excluding employers of the same type as the organization.

5 . The computer-implemented method of claim 1 , wherein the determining a feature set further comprises:
de-duplicating features in the feature set.

6 . The computer-implemented method of claim 1 , wherein the determining a feature set further comprises:
determining that a feature value is likely false based on a veracity score that does not satisfy a threshold veracity value; and re-labeling the feature value.

7 . The computer-implemented method of claim 1 , wherein the providing a ranking of users further comprises:
sorting the users in the evaluation set based on associated scores determined by the model; adjusting the scores for users who are current employees or previous employees of the organization; and generating an ordered list of scores and associated users.

8 . The computer-implemented method of claim 7 , further comprising:
creating a look up table to maintain the ordered list of scores and associated users.

9 . The computer-implemented method of claim 8 , further comprising:
in response to an indication provided by a particular user through a user interface to identify and rank job candidates for the organization, selecting connections of the particular user and their associated scores from the look up table.

10 . The computer-implemented method of claim 9 , further comprising:
presenting through the user interface the selected connections in an order based on their associated scores.

11 . 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: determining a feature set for the model; training the model based on the training set and the feature set to determine a score reflecting a probability that each user in an evaluation set of users is qualified for employment with an organization; and providing a ranking of users in the evaluation set based on the score determined for each user.

12 . The system of claim 11 , wherein the training set includes users who are employees of the organization.

13 . The system of claim 12 , wherein the training set further includes users who are employees of a second organization of the same organization type as the organization.

14 . The system of claim 11 , wherein the feature set comprises at least one of a number of connections of a user in the training set on a social networking system, a number of days since the user performed an action on the social networking system, a number of requests by the user to initiate connections on the social networking system, a number of entities who are following the user on the social networking system, a number of entities followed by the user on the social networking system, college attended by the user, graduate school attended by the user, degrees obtained by the user, concentrations of study by the user, and employers of the user excluding employers of the same type as the organization.

15 . The system of claim 11 , wherein the determining a feature set further comprises:
de-duplicating features in the feature set.

16 . 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:
determining a feature set for the model; training the model based on the training set and the feature set to determine a score reflecting a probability that each user in an evaluation set of users is qualified for employment with an organization; and providing a ranking of users in the evaluation set based on the score determined for each user.

17 . The non-transitory computer-readable storage medium of claim 16 , wherein the training set includes users who are employees of the organization.

18 . The non-transitory computer-readable storage medium of claim 17 , wherein the training set further includes users who are employees of a second organization of the same organization type as the organization.

19 . The non-transitory computer-readable storage medium of claim 16 , wherein the feature set comprises at least one of a number of connections of a user in the training set on a social networking system, a number of days since the user performed an action on the social networking system, a number of requests by the user to initiate connections on the social networking system, a number of entities who are following the user on the social networking system, a number of entities followed by the user on the social networking system, college attended by the user, graduate school attended by the user, degrees obtained by the user, concentrations of study by the user, and employers of the user excluding employers of the same type as the organization.

20 . The non-transitory computer-readable storage medium of claim 16 , wherein the determining a feature set further comprises:
de-duplicating features in the feature set.

Assignees

• Facebook Inc

Inventors

• Fang Miaoqing

Classifications

• G06F16/24578

  using ranking · CPC title

• G06Q10/105

  Human resources · CPC title

• G06F16/24556

  Aggregation; Duplicate elimination · CPC title

• G06F17/3053

  Physics · mapped topic

• G06F17/30489

  Physics · mapped topic