Abstract: Described are systems and methods for generating a debiased training dataset and training one or more stages of a multi-stage recommendation system and/or service using the debiased training dataset. Rather than generating training datasets from the output data generated by the multi-stage recommendation system and/or data served by later stages of the multi-stage recommendation system, debiased training datasets may be generated from data that was served by the particular stage of the multi-stage recommendation system that is to be trained using the generated debiased training dataset. The debiased training dataset may be generated by generating pseudo-labels for each data record and comparing the generated pseudo-labels against two thresholds to generate a binary classification of the data records.

Abstract: A notification platform for distribution of notification content in an on-line social network system, on-line and in near real time, is described. As a new notification is detected in the continuous stream of notifications triggered by time-sensitive events, the near real time notifications distribution system determines member profiles representing potential recipients of the notification by traversing a relationship graph. The relationship graph has nodes representing member profiles, as well as other entities maintained in the on-line social network system. The edges of the relationship graph represent relationships between entities represented by the associated nodes. For each member profile representing a potential recipient of the notification, the near real time notifications distribution system generates a relevance score, which is used to determine whether the notification is to be delivered to the potential recipient.

Abstract: A notification platform for distribution of notification content in an on-line social network system, on-line and in near real time, is described. As a new notification is detected in the continuous stream of notifications triggered by time-sensitive events, the near real time notifications distribution system determines member profiles representing potential recipients of the notification by traversing a relationship graph. The relationship graph has nodes representing member profiles, as well as other entities maintained in the on-line social network system. The edges of the relationship graph represent relationships between entities represented by the associated nodes. For each member profile representing a potential recipient of the notification, the near real time notifications distribution system generates a relevance score, which is used to determine whether the notification is to be delivered to the potential recipient.