Systems and methods for recommending job postings
Systems, methods, and non-transitory computer-readable media can determine a user embedding for a user of a content provider based on one or more machine learning models. A job embedding for a job posting published through the content provider can be determined based on one or more machine learning models. A determination can be made of a threshold level of similarity between the user embedding and the job embedding. The job posting can be provided as a recommendation to the user through the content provider.
1. A computer-implemented method comprising:
applying, by a computing system, one or more machine learning models to generate a user embedding for a user of a content provider, the user embedding based at least in part on i) a user profile of the user and ii) content associated with an organization that was accessed by the user;
applying, by the computing system, one or more machine learning models to generate a job embedding for a job posting published through the content provider, wherein
the job embedding comprises at least a first portion corresponding to a text embedding, a second portion corresponding to a media embedding, and a third portion corresponding to a data embedding,
the media embedding is generated based at least in part on media included in the job posting, the media including an image, and
the data embedding is generated based at least in part on a number of times the job posting was accessed by users, a number of times the job posting was shared by users, and a number of users that have applied to the job posting;
determining, by the computing system, a threshold level of similarity between the user embedding and the job embedding; and
providing, by the computing system, the job posting as a recommendation to the user through the content provider.
2. The computer-implemented method of claim 1 , wherein the user embedding for the user comprises at least a first portion corresponding to a text embedding, a second portion corresponding to a media embedding, and a third portion corresponding to a data embedding.
3. The computer-implemented method of claim 2 , wherein the text embedding of the user embedding is generated based at least in part on text included in a user profile of the user and the text included in content that was accessed by the user.
4. The computer-implemented method of claim 2 , wherein the media embedding of the user embedding is generated based at least in part on media included in a user profile of the user, media included in content that was accessed by the user, or both.
5. The computer-implemented method of claim 2 , wherein the data embedding of the user embedding is generated based at least in part on a set of identifiers corresponding to content items that were sequentially accessed by the user over one or more sessions.
6. The computer-implemented method of claim 1 , wherein the text embedding of the job embedding is generated based at least in part on text included in the job posting.
7. The computer-implemented method of claim 1 , wherein determining the threshold level of similarity between the user embedding and the job embedding further comprises:
determining, by the computing system, a similarity score based on a dot product of the user embedding and the job embedding; and
determining, by the computing system, that the similarity score satisfies a threshold value.
8. The computer-implemented method of claim 1 , wherein the text embedding of the user embedding is generated by a first machine learning model, the media embedding of the user embedding is generated by a second machine learning model, and the data embedding of the user embedding is generated by a third machine learning model, the third machine learning model including a recurrent neural network layer for processing sequentially ordered data.
9. 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:
applying one or more machine to generate a user embedding for a user of a content provider, the user embedding based at least in part on i) a user profile of the user and ii) content associated with an organization that was accessed by the user;
applying one or more machine learning models to generate a job embedding for a job posting published through the content provider, wherein
the job embedding comprises at least a first portion corresponding to a text embedding, a second portion corresponding to a media embedding, and a third portion corresponding to a data embedding,
the media embedding is generated based at least in part on media included in the job posting, the media including an image, and
the data embedding is generated based at least in part on a number of times the job posting was accessed by users, a number of times the job posting was shared by users, and a number of users that have applied to the job posting;
determining a threshold level of similarity between the user embedding and the job embedding; and
providing the job posting as a recommendation to the user through the content provider.
10. The system of claim 9 , wherein the user embedding for the user comprises at least a first portion corresponding to a text embedding, a second portion corresponding to a media embedding, and a third portion corresponding to a data embedding.
11. The system of claim 10 , wherein the text embedding of the user embedding is generated based at least in part on text included in a user profile of the user and the text included in content that was accessed by the user.
12. The system of claim 10 , wherein the media embedding of the user embedding is generated based at least in part on media included in a user profile of the user, media included in content that was accessed by the user, or both.
13. The system of claim 10 , wherein the data embedding of the user embedding is generated based at least in part on a set of identifiers corresponding to content items that were sequentially accessed by the user over one or more sessions.
14. The system of claim 9 , wherein the text embedding of the user embedding is generated by a first machine learning model, the media embedding of the user embedding is generated by a second machine learning model, and the data embedding of the user embedding is generated by a third machine learning model, the third machine learning model including a recurrent neural network layer for processing sequentially ordered data.
15. 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:
applying one or more machine learning models to generate a user embedding for a user of a content provider, the user embedding based at least in part i) a user profile of the user and ii) content associated with an organization that was accessed by the user;
applying one or more machine learning models to generate a job embedding for a job posting published through the content provider, wherein
the job embedding comprises at least a first portion corresponding to a text embedding, a second portion corresponding to a media embedding, and a third portion corresponding to a data embedding,
the media embedding is generated based at least in part on media included in the job posting, the media including an image, and
the data embedding is generated based at least in part on a number of times the job posting was accessed by users, a number of times the job posting was shared by users, and a number of users that have applied to the job posting;
determining a threshold level of similarity between the user embedding and the job embedding; and
providing the job posting as a recommendation to the user through the content provider.
16. The non-transitory computer-readable storage medium of claim 15 , wherein the user embedding for the user comprises at least a first portion corresponding to a text embedding, a second portion corresponding to a media embedding, and a third portion corresponding to a data embedding.
17. The non-transitory computer-readable storage medium of claim 16 , wherein the text embedding of the user embedding is generated based at least in part on text included in a user profile of the user and the text included in content that was accessed by the user.
18. The non-transitory computer-readable storage medium of claim 16 , wherein the media embedding of the user embedding is generated based at least in part on media included in a user profile of the user, media included in content that was accessed by the user, or both.
19. The non-transitory computer-readable storage medium of claim 16 , wherein the data embedding of the user embedding is generated based at least in part on a set of identifiers corresponding to content items that were sequentially accessed by the user over one or more sessions.
20. The non-transitory computer-readable storage medium of claim 15 , wherein the text embedding of the user embedding is generated by a first machine learning model, the media embedding of the user embedding is generated by a second machine learning model, and the data embedding of the user embedding is generated by a third machine learning model, the third machine learning model including a recurrent neural network layer for processing sequentially ordered data.