Abstract: A system and method for inferring appropriate courses for recommendation based on member characteristics is disclosed. A social networking system receives a request for recommended courses, wherein the request is associated with a member of the social networking system. The social networking system identifies a group of members who are similar to the first member. The social networking system creates a list of recently learned skills by members of the group of members similar to the member. For a particular skill in the list of skills, the social networking system determines whether the member possesses the particular skill. In accordance with a determination that the member does not possess the particular skill, the social networking system identifies at least one course that teaches the particular skill from a list of courses. The social networking system transmits the identified course to the client device for display as a recommended course.

Abstract: Techniques for modifying training data for video response quality optimization are provided. In one technique, training data is identified that is generated based on video presentation data that indicates multiple video items were presented to multiple entities. The training data comprises multiple training instances, each indicating a presentation of at least a portion of a video item to an entity. For each training instance in a subset of the training instances, a quality metric of the presentation of the video item indicated in said each training instance is computed and that training instance is modified based on the quality metric. After modifying one or more of the training instances, the model is trained using one or more machine learning techniques. In response to a content request, the model is used to determine whether to transmit a particular video item over a network to a computing device of a particular entity.

Abstract: Techniques for modifying training data for video response quality optimization are provided. In one technique, training data is identified that is generated based on video presentation data that indicates multiple video items were presented to multiple entities. The training data comprises multiple training instances, each indicating a presentation of at least a portion of a video item to an entity. For each training instance in a subset of the training instances, a quality metric of the presentation of the video item indicated in said each training instance is computed and that training instance is modified based on the quality metric. After modifying one or more of the training instances, the model is trained using one or more machine learning techniques. In response to a content request, the model is used to determine whether to transmit a particular video item over a network to a computing device of a particular entity.

Abstract: Techniques for inferring a specific skill associated with a job posting are described. In an example, disclosed is a system that selects, from a jobs database, a specific job posting from a plurality of job postings. Additionally, job applicants for the specific job posting can be determined using indicators in the profile data of members. Moreover, a set of skills associated with the job applicants can be obtained. Furthermore, a percentage of the job applicants having a specific skill from the set of skills can be determined using the profile data of the job applicants. Subsequently, a confidence score of the specific skill being associated with the specific job posting can be calculated based on the percentage of the job applicants having the specific skill. A user interface can display a presentation of the specific job posting to a first member when the confidence score transgresses a predetermined score.

Abstract: Techniques for learning and leveraging embeddings for response prediction are provided. Based on training data, an embedding for each attribute value of multiple content items is generated, an embedding for each attribute value of multiple entities is generated, weights of a first neural network for content items is generated, and weights of a second neural network for requesting entities is generated. In response to receiving a request, a particular content item is identified. A first set of embeddings for the particular content item is identified and input into the first neural network to generate first output. A particular requesting entity that initiated the content request is identified. A second set of embeddings for the particular requesting entity is identified and input into the second neural network to generate second output. The particular content item is selected based on the first output and the second output.

Abstract: Techniques for improving accuracy of data storage and retrieval using a parser for dynamically updating data for storage are disclosed herein. In some embodiments, a method comprises: detecting an event triggered by a user input on a device; receiving a company, a title, and a description for a posting, the company identification, the title, and the description being receiving in association with the event; generating a list of skills based on the company identification, the title, and the description, the generating the list of skills comprising: identifying a first plurality of skills based on a search using the company identification and the title, identifying a second plurality of skills based on a parsing of the description, and generating the list of skills based on the first plurality of skills and the second plurality of skills; and displaying the generated list of skills to on the device.

Abstract: Systems and methods for identifying appropriate course recommendations are disclosed. A system receives a request for recommended courses and accesses a plurality of course records to determine whether a sufficient number of courses that teach the skill of interest are available. In accordance with a determination that a sufficient number of courses that teach the skill of interest are not available, the system generates skill attribute vectors for a skill of interest and a plurality of other skills and ranks the plurality of other skills based on the distance between the skill attribute for the skill of interest and the plurality of other skills. The system selects a skill based on the rankings. The system identifies at least one course that teaches the selected skill and transmits a course recommendation for the identified course to a client system for presentation.

Abstract: A system and method for inferring appropriate courses for recommendation based on member characteristics is disclosed. A social networking system receives a request for recommended courses, wherein the request is associated with a member of the social networking system. The social networking system identifies a group of members who are similar to the first member. The social networking system creates a list of recently learned skills by members of the group of members similar to the member. For a particular skill in the list of skills, the social networking system determines whether the member possesses the particular skill. In accordance with a determination that the member does not possess the particular skill, the social networking system identifies at least one course that teaches the particular skill from a list of courses. The social networking system transmits the identified course to the client device for display as a recommended course.

Abstract: Techniques for inferring a specific skill associated with a job posting are described. In an example, disclosed is a system that selects, from a jobs database, a specific job posting from a plurality of job postings. Additionally, job applicants for the specific job posting can be determined using indicators in the profile data of members. Moreover, a set of skills associated with the job applicants can be obtained. Furthermore, a percentage of the job applicants having a specific skill from the set of skills can be determined using the profile data of the job applicants. Subsequently, a confidence score of the specific skill being associated with the specific job posting can be calculated based on the percentage of the job applicants having the specific skill. A user interface can display a presentation of the specific job posting to a first member when the confidence score transgresses a predetermined score.

Abstract: Techniques for associating a specific skill with a job posting are described. In an example, disclosed is a system having a standardization database and a skills database. The system can extract textual features from a job posting, and select a specific skill for the job posting from a plurality of skills. Moreover, a job title for the job posting can be selected from a plurality of standardized job titles. Furthermore, an affinity score can be accessed for specific skill based on the selected job title. The system can determine a skill frequency score associated with the specific skill. The system can calculate a confidence score of the specific skill based on the textual features, the affinity score, and the skill frequency score. Subsequently, the system can store, in the skills database, an association of the specific skill with the job posting when the confidence score is above a predetermined threshold.