Abstract: 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.

Abstract: Systems, methods, and non-transitory computer readable media are configured to generate multiple channel embeddings for a page of a social networking system. The multiple channel embeddings can be mapped to a shared embedding space. A page embedding for the page of the social networking system can then be generated.

Abstract: Systems, methods, and non-transitory computer-readable media can receive page information associated with a page and user information associated with a user associated with the page. Confidence scores are calculated for a plurality of categories based on the page information and the user information, wherein a confidence score for a category is indicative of a likelihood that the category is relevant to the page. One or more categories of the plurality of categories are selected based on the confidence scores. The one or more categories are presented to the user as category recommendations.

Abstract: Exemplary embodiments relate to techniques for identifying job listing posts in a platform, such as a social networking or messaging service. Job listing posts can be identified as they are created, causing the user to enter a job posting interface for entering structured data. The structured data may be searchable on the platform. Job posts can also be identified post hoc, and subsequently converted to structured job listing posts. Structured information may be drawn from the freeform text, and the system may normalize the job description (e.g., using third-party information, such as standard job descriptors). Identifying a job listing post/intent may be done using a model trained using various parameters, such as: user feedback/administrator actions; corrections; content of the post; existing posts of page owner; the time of year; whether other employers in similar fields are post job openings; and whether this owner prefers to write structured or unstructured posts.

Abstract: Exemplary embodiments relate to techniques for identifying job seekers in a platform, such as a social networking or messaging service, and presenting the job seekers with job postings that they are most likely to be interested in. The job seeker's intent may be determined based on various factors. The system may rank job postings by the probability that a user will be interested in a particular job, and can surface jobs to the user in a number of ways. Applicable jobs can be surfaced in an interactable list.

Abstract: Systems, methods, and non-transitory computer-readable media can receive page information associated with a page and user information associated with a user associated with the page. Confidence scores are calculated for a plurality of categories based on the page information and the user information, wherein a confidence score for a category is indicative of a likelihood that the category is relevant to the page. One or more categories of the plurality of categories are selected based on the confidence scores. The one or more categories are presented to the user as category recommendations.