Abstract: This disclosure is directed to determining various economic graph indices and, in particular, to systems and methods that leverage a graph analytic engine and framework to determine values assigned to graph nodes extracted from one or more member profiles, and visualizing said values to correlate skills, geographies, and industries. The disclosed embodiments include a client-server architecture where a social networking server has access to a social graph of its social networking members. The social networking server includes various modules and engines that import the member profiles and then extracts certain defined attributes from the member profiles, such as employer (e.g., current employer and/or past employers), identified skills, educational institutions attended, and other such defined attributes. Using these attributes as nodes, the social networking server constructs a graph using various graph processing techniques.

Abstract: Apparatuses, computer readable media, and methods are disclosed for generating machine learned models and recommendations identified using hidden feature vectors determined using the machine learned models. The method includes selecting a job profile associated with a first set of members of a social networking system. The method identifies a set of interactions with the job profile, where the set of actions are taken by a second set of members, and generates a vector model for the job profile. The vector model identifies a set of hidden feature vectors for the job profile. The method determines a job recommendation based on the job profile, the set of interactions, the set of second members, and the vector profile. The method then causes presentation of the job recommendation on a display device of a computing device.

Abstract: Apparatuses, computer-readable medium, and methods are disclosed for generating machine learned models and recommendations identified using hidden feature vectors determined using the machine learned models. The method includes selecting a first job profile associated with a first set of members of a social networking system. The method identifies at least one second job profile associated with a second set of members and generates a vector model for the first job profile. The vector model identifies a set of hidden feature vectors for the first job profile. The method determines a job recommendation based on the first job profile, the at least one second job profile, the vector model, and a selected member profile. The method then causes presentation of the job recommendation on a display device of a computing device associated with the selected member profile.

Abstract: Apparatuses, computer readable medium, and methods are disclosed for generating job recommendations. The method includes determining one or more first job profiles that are similar to a second job profile of a member of a social network system, and determining first regression coefficients and a first hidden feature vector jointly for a first layer of a hierarchical structure based on the one or more first job profiles and the second job profile. The method may further include determining one or more third job profiles that are similar to the second job profile, wherein the one or more third job profiles are from a same company as the second job profile, and determining second regression coefficients and a second hidden feature vector jointly for a second layer of the hierarchical structure based on the first regression coefficients, the first hidden feature vector, and the one or more third job profiles.

Abstract: Apparatuses, computer readable medium, and methods are disclosed for generating job recommendations using a hierarchical Bayesian structure. The method of generating job recommendations includes determining, by at least one hardware processor, first regression coefficients and first hidden feature vector jointly for a first layer based on a member's view behavior, and the member's profile. The method further includes determining, by the at least one hardware processor, second regression coefficients and second hidden feature vector jointly for a second layer based on the first regression coefficients, the first hidden feature vector, and the member's application behavior. The method further includes determining, by the at least one hardware processor, a job recommendation based on one or more job profiles, the first regression coefficients, first hidden feature vector, second regression coefficients, and second hidden feature vector.

Abstract: A system and method for marginal value-based content item mixing are provided. In example embodiments, a request to fill a position within a content list is received. A variable monetary value for positioning a particular job listing within the content list is calculated based on an average monetary value and an incremental value for the particular job listing. A first expected value for the particular job listing is computed based on the variable monetary value and an interaction likelihood for the particular job listing. The first expected value is compared with a second expected value corresponding to filling the position within the content list with another content item. The particular job listing within the content list is presented on a user interface of a user device based on the first expected value exceeding the second expected value.

Abstract: A JPV calculator is described for quantifying a job poster value for a particular job posting at a certain point in time, in the context of an on-line social network system. A job poster value (JPV) represents a value that is still owed to the associated job posting entity at a certain point in time with respect to a particular job posting. The JPV calculator determines JPV by taking into account the temporal job statistics for a job. The job poster value drops as the job posting gets more applications from members of the on-line social network system.

Abstract: Learning to rank modeling in the context of an on-line social network is described. A learning to rank model can learn from pairwise preference (e.g., job posting A is more relevant than job posting B for a particular member profile) thus directly optimizing for the rank order of job postings for each member profile. With ranking position taken into consideration during training, top-ranked job postings may be treated by a recommendation system as being of more importance than lower-ranked job postings. In addition, a learning to rank approach may also result in an equal optimization across all member profiles and help minimize bias towards those member profiles that have been paired with a larger number of job postings.

Abstract: System and method are described for determining whether a job posting is to be presented to a member of an on-line social network system. A JPV calculator determines a job poster value for the job posting at a certain point in time, in the context of the on-line social network system. A job poster value (JPV) represents a value that is still owed to the associated job posting entity at a certain point in time with respect to a particular job posting. A relevance value calculator determines relevance value for a job posting based on results of comparing features of a member profile that represents the member and respective features of the job posting. A determination of whether the job posting is to be presented to the member is based on both its relevance value and its JPV.

Abstract: System and method are described for determining an aggregated job poster value (JPV) for a job posting entity in an on-line social network system. JPV represents a value that is still owed to the associated job posting entity at a certain point in time with respect to a particular job posting. An aggregated job poster value is determined using an aggregation function that aggregates job poster values for all job postings that are associated with a particular job posting entity and that satisfy certain predetermined criteria.

Abstract: A system, a machine-readable storage medium storing instructions, and a computer-implemented method are described herein are directed to a Term Weight Engine that defines a pairing comprising a user profile text section paired with a job post text section. The Term Weight Engine learns a pairing weight indicating an extent that a similarity of text in the pairing predicts a relevance of a respective job posting to a given user profile. The Term Weight Engine learns a global weight for a term(s). The Term Weight Engine calculates a similarity score of the pairing as between a first user profile of a target member account and a first job posting. Based on identifying the term appears in the pairing as between a first user profile of a target member account and a first job posting, the Term Weight Engine applies the global weight to the similarity score to generate a prediction indicating whether the target member account will apply to the first job posting.

Abstract: Techniques for presenting a personalized job posting are described. A job publisher can access member data of a first member from a member database and a plurality of job postings from a job database. A job score for each job posting in the plurality of job postings can be calculated based on the profile data of the first member and the job data for each job posting. Additionally, a relevant job posting for the first member can be determined based on the calculated job score for each job posting in the plurality of job postings. Moreover, a second member for the relevant job posting can be selected based the social graph data of the first member and the job data for the relevant job posting. Furthermore, a presentation of the relevant job posting can be on a display of a device of the first member.

Abstract: Learning to rank modeling in the context of an on-line social network is described. A learning to rank model can learn from pairwise preference (e.g., job posting A is more relevant than job posting B for a particular member profile) thus directly optimizing for the rank order of job postings for each member profile. With ranking position taken into consideration during training, top-ranked job postings may be treated by a recommendation system as being of more importance than lower-ranked job postings. In addition, a learning to rank approach may also result in an equal optimization across all member profiles and help minimize bias towards those member profiles that have been paired with a larger number of job postings.

Abstract: This disclosure is directed to determining various economic graph indices and, in particular, to systems and methods that leverage a graph analytic engine and framework to determine values assigned to graph nodes extracted from one or more member profiles, and visualizing said values to correlate skills, geographies, and industries. The disclosed embodiments include a client-server architecture where a social networking server has access to a social graph of its social networking members. The social networking server includes various modules and engines that import the member profiles and then extracts certain defined attributes from the member profiles, such as employer (e.g., current employer and/or past employers), identified skills, educational institutions attended, and other such defined attributes. Using these attributes as nodes, the social networking server constructs a graph using various graph processing techniques.

Abstract: Learning to rank modeling in the context of an on-line social network is described. A learning to rank model can learn from pairwise preference (e.g., job posting A is more relevant than job posting B for a particular member profile) thus directly optimizing for the rank order of job postings for each member profile. With ranking position taken into consideration during training, top-ranked job postings may be treated by a recommendation system as being of more importance than lower-ranked job postings. In addition, a learning to rank approach may also result in an equal optimization across all member profiles and help minimize bias towards those member profiles that have been paired with a larger number of job postings.

Abstract: Nonlinear featurization of decision trees for linear regression modeling in the context of an on-line social network is described. A computer-implemented converter is provided that is capable of reading a decision tree structure that is included in the learning to rank algorithm and convert each path from root to a leaf into an s-expression. The s-expressions are used as additional features to train a logistic regression model.

Abstract: A system stores member and job posting pairs in a database, which indicate that a job has been presented to a member, that the member has viewed the job posting, and that the member has applied for the job. The system extracts member attributes from a member profile and job attributes from a job posting file, and creates attribute pairs from the extracted member attributes and the extracted job attributes. The system creates an attribute tracking table, and for each job attribute in the attribute tracking table, create ratios relating to jobs that have been shown to members, jobs that members have viewed, and jobs for which members have applied. The system also determines trends between particular member attributes and particular job attributes. The ratios and trends are provided to a job recommendation engine that uses them in determining jobs to recommend to members.

Abstract: Techniques for standardizing and deduplicating unpaid job postings obtained from third-party systems are described. An unpaid job posting is obtained by a social networking service from a third-party system. The title and description of the unpaid job posting are standardized and combined into a standardized unpaid job posting. A deduplication process is performed to prevent the standardized unpaid job posting from replacing a paid job posting within the social networking service, and to prevent the standardized unpaid job posting from replacing a more authoritative, unpaid job posting within the social networking service.

Abstract: Generally discussed herein are methods, systems, and apparatuses for determining a latent preference of a user. One or more embodiments, discussed herein regard determining a latent preference of a user's propensity to relocate for a job. According to an example, a method can include receiving one or more characteristics of a user of a web service, estimating a probability corresponding to a latent preference of the user, and/or determining whether the probability indicates the user has the latent preference.

Abstract: A computer-based method, and computer system, for matching candidates with job openings. The technology more particularly relates to methods of providing a candidate with a score for a particular job opening, where the score is derived from a comparison of features in the candidate's resume with job features in a description of the job opening, as well as use of external data gathered from other sources and based on information contained in the candidate's resume and/or in the description of the job opening. Particular features are weighted to take account of their significance in matching candidates to job openings in a statistical survey of such matching. The technology further provides for notifying employers that one or more high scoring candidates have been identified.