Abstract: Techniques for a social networking system involve analyzing members' interactions to enhance follow recommendations. The techniques include calculating the number of follows generated by a first member and the number of follows received by a second and third member. The platform then computes a weighted follow utility score for two member pairs: the first comprising the first and second members, and the second comprising the first and third members. These scores are determined by equally weighing the follows generated and received, along with a probability of the first member following the second or third member. While the probability suggests a preference for the second member, the ranking system places the third member pair higher. Consequently, the first member is advised to follow the third member, based on this ranking. Finally, this recommendation is displayed to the first member on an electronic device, guiding their social interactions on the platform.

Abstract: Embodiments of the disclosed technologies include generating a reward score for an entity. A rate distribution is determined using the reward score and a number of times the entity has been selected for ranking. A sampled rate value is generated by sampling the rate distribution. A probability score is generated for a pair of the entity and a user based on the sampled rate value. A probability distribution is determined using the probability score. A sampled probability value is generated by sampling the probability distribution. A machine learning model is trained using the sampled probability value.

Abstract: Embodiments of the disclosed technologies include generating a reward score for an entity. A rate distribution is determined using the reward score. A sampled rate value is generated by sampling the rate distribution. A probability score is generated for a pair of the entity and a user using the sampled rate value. A probability distribution is determined using the probability score. A sampled probability value is generated by sampling the probability distribution. A machine learning model is trained using the sampled probability value.

Abstract: Embodiments of the disclosed technologies include generating a reward score for an entity. A rate distribution is determined using the reward score and a number of times the entity has been selected for ranking. A sampled rate value is generated by sampling the rate distribution. A probability score is generated for a pair of the entity and a user based on the sampled rate value. A probability distribution is determined using the probability score. A sampled probability value is generated by sampling the probability distribution. A machine learning model is trained using the sampled probability value.

Abstract: Embodiments of the disclosed technologies include generating a reward score for an entity. A rate distribution is determined using the reward score. A sampled rate value is generated by sampling the rate distribution. A probability score is generated for a pair of the entity and a user using the sampled rate value. A probability distribution is determined using the probability score. A sampled probability value is generated by sampling the probability distribution. A machine learning model is trained using the sampled probability value.

Abstract: In an example embodiment, a blending model is presented based on a linear programming approach. The blending model produces a slate of sponsored and non-sponsored pieces of content for display in a graphical user interface, with the ordering and placement of the sponsored and non-sponsored pieces of content selected in order to maximize an objective function. Such an approach can fine tune each piece of content using content-level parameters and holistically examine global constraints and opportunities. It establishes a robust optimization framework that can adapt to content and domain changes without requiring tuning through online experiments.

Abstract: Sponsored and organic pieces of content are displayed in accordance with a blending model that is used to first identify a pattern of slots to which to assign either sponsored or organic pieces of content. This blending model is applied to a combination of both sponsored and non-sponsored pieces of content being considered for display to a user. This pattern only determines the slot assignments. The actual ranking of the pieces of content, and more particularly the actual ranking of the organic pieces of content, is determined by an ordering other than the ranking determined by the blending model, such as by using the original ordering of the second list. The pieces of content are then displayed in the order of this actual ranking, but in the slots indicated as having been assigned to be either sponsored or organic in the pattern determined by the blending model.

Abstract: Sponsored and organic pieces of content are displayed in accordance with a blending model that is used to first identify a pattern of slots to which to assign either sponsored or organic pieces of content. This blending model is applied to a combination of both sponsored and non-sponsored pieces of content being considered for display to a user. This pattern only determines the slot assignments. The actual ranking of the pieces of content, and more particularly the actual ranking of the organic pieces of content, is determined by an ordering other than the ranking determined by the blending model, such as by using the original ordering of the second list. The pieces of content are then displayed in the order of this actual ranking, but in the slots indicated as having been assigned to be either sponsored or organic in the pattern determined by the blending model.

Abstract: Methods, systems, and computer programs are presented for presenting return-on-investment (ROI) information, for budgeted services that resulted in a successful service delivery, on a user interface for setting the budget for a service request. One method includes an operation for identifying daily budgets for budgeted services that resulted in a successful service delivery (BSSSD). Each daily budget indicates an amount for spending in promotion of the BSSSD in an online service. The method further includes receiving a request, in a graphical user interface (GUI) of the online service, for posting a daily budget for a first budgeted service. Further, a performance value, associated with the daily budgets of the BSSSD that are similar to the first budgeted service, is selected. Further, the method includes causing presentation, by the one or more processors, of the performance value in the GUI.

Abstract: In an example embodiment, a bid of an impression of a piece of content, while dynamically set at impression time, may be based on a base bid that is something of a rough indicator of what the estimated price will be. That base bid then is adjusted dynamically up or down at impression time. This base bid can be determined by dividing the expected number of impressions for a day by a total daily budget. The expected number of impressions may be determined by using the empirical number of impressions from the previous day. As such, in an example embodiment, the prediction of the number of impressions for a day utilizes a corrected version of the empirical number of impressions from the prior day, with the corrected version based on a specialized formula with weights trained by a machine learning algorithm.

Abstract: Techniques for the dynamic slotting of content items within electronic content are provided. In response to receiving a request, a first set of content items is identified and a first score is generated for each based on a first objective. Based on the first scores, a first ranking of the first set of content items is generated. A subset of the first set is selected based on the first ranking. A second set of content items that includes the subset is identified and a second score is generated for each item in the second set based on a second objective that is different than the first objective. Based on the second scores, a second ranking of the second set is generated. A subset of the second set is selected based on the second ranking. The second subset is presented on a computing device that is associated with the request.

Abstract: Techniques for predicting relevance of social networking service member accounts to a job posting. In an embodiment, a candidate predictor engine of a system encodes data representing an applicant quality (AQ) score for each job/applicant pair for a plurality of applicants to a job posting. Additionally, the system stores the encoded data and assigns member-level weights to each of the applicants. Moreover, the system calculates weighted AQ scores for each of the job/applicant pairs, the weighted AQ scores being products of respective AQ scores and member-level weights. Furthermore, the system sums the weighted AQ scores to derive a total weighted score for the job posting. Then, the candidate predictor engine generates a job-level probability of confirmed hire (pCH) based on the total weighted score, the job-level pCH indicating a likelihood of the posting being filled by an applicant. Also, the system transmits the job-level pCH to a client for display.

Abstract: The disclosed embodiments provide a system for performing dynamic job bidding optimization. During operation, the system obtains historical data containing a time series of interactions with a job. Next, the system uses the historical data to calculate an initial price of a job based on a predicted number of interactions with the job. The system then determines a first dynamic adjustment to the initial price that improves utilization of a budget for the job and a second dynamic adjustment to the initial price that improves a performance of the job. Finally, the system applies the first and second adjustments to the initial price to produce an updated price for the job and delivers the job within an online system based on the updated price.

Abstract: Techniques for the dynamic slotting of content items within electronic content are provided. In response to receiving a request, a first set of content items is identified and a first score is generated for each based on a first objective. Based on the first scores, a first ranking of the first set of content items is generated. A subset of the first set is selected based on the first ranking. A second set of content items that includes the subset is identified and a second score is generated for each item in the second set based on a second objective that is different than the first objective. Based on the second scores, a second ranking of the second set is generated. A subset of the second set is selected based on the second ranking. The second subset is presented on a computing device that is associated with the request.

Abstract: The disclosed embodiments provide a system that performs pacing for balanced delivery. During operation, the system obtains predicted response rates associated with impressions of a content item delivered within an online system and a cost per action (CPA) for the content item. Next, the system determines an impression-based spending for the content item based on the predicted response rates and the CPA. The system then calculates a pacing score for the content item based on the impression-based spending. Finally, the system adjusts subsequent interactions with the content item based on the pacing score.

Abstract: The disclosed embodiments provide a system for performing an experiment strategy for member-based ramping. During operation, the system divides members of an online system into a control group and a treatment group. Next, the system configures delivery of content to the control group over a first set of channels that charge for actions related to the content and delivery of the content to the treatment group over a second set of channels that charge for the actions related to the content, wherein the second set of channels is smaller than the first set of channels. The system then performs a first experiment that selects an adjustment factor for content delivered over the second set of channels to the treatment group to achieve revenue neutrality and/or engagement neutrality between the treatment and control groups.

Abstract: In an example embodiment, a bid of an impression of a piece of content, while dynamically set at impression time, may be based on a base bid that is something of a rough indicator of what the estimated price will be. That base bid then is adjusted dynamically up or down at impression time. This base bid can be determined by dividing the expected number of impressions for a day by a total daily budget. The expected number of impressions may be determined by using the empirical number of impressions from the previous day. As such, in an example embodiment, the prediction of the number of impressions for a day utilizes a corrected version of the empirical number of impressions from the prior day, with the corrected version based on a specialized formula with weights trained by a machine learning algorithm.

Abstract: Methods, systems, and computer programs are presented for presenting return-on-investment (ROI) information, for budgeted services that resulted in a successful service delivery, on a user interface for setting the budget for a service request. One method includes an operation for identifying daily budgets for budgeted services that resulted in a successful service delivery (BSSSD). Each daily budget indicates an amount for spending in promotion of the BSSSD in an online service. The method further includes receiving a request, in a graphical user interface (GUI) of the online service, for posting a daily budget for a first budgeted service. Further, a performance value, associated with the daily budgets of the BSSSD that are similar to the first budgeted service, is selected. Further, the method includes causing presentation, by the one or more processors, of the performance value in the GUI.

Abstract: The disclosed embodiments provide a system for performing dynamic job bidding optimization. During operation, the system obtains historical data containing a time series of interactions with a job. Next, the system uses the historical data to calculate an initial price of a job based on a predicted number of interactions with the job. The system then determines a first dynamic adjustment to the initial price that improves utilization of a budget for the job and a second dynamic adjustment to the initial price that improves a performance of the job. Finally, the system applies the first and second adjustments to the initial price to produce an updated price for the job and delivers the job within an online system based on the updated price.

Abstract: The disclosed embodiments provide a system for performing delivery-based optimizations for jobs. During operation, the system obtains historical data comprising delivery rates related to overdelivery and underdelivery of jobs with respect to fixed budgets for the jobs. Next, the system applies an optimization technique to the historical data to determine a change to an initial price for a job that improves a utilization of a fixed budget for the job. The system then applies the change to the initial price to produce an updated price for the job and delivers the job within an online system based on the updated price.