PERSONALIZED DELIVERY TIME OPTIMIZATION

摘要

Techniques for optimizing a delivery time for the delivery of messages are described. According to various embodiments, a system determines, for each of a plurality of time intervals, a likelihood of a particular member of an online social network service performing a particular member user action on a particular message content item during the corresponding time interval. The plurality of time intervals are then ranked, based on the determined likelihoods corresponding to the plurality of time intervals. Thereafter, a particular time interval is identified from among the plurality of time intervals that is associated with a highest ranking. The particular time interval is then classified as an optimum personalized message delivery time for the particular member.

主权项

1. A computer-implemented method comprising:
determining, by a machine including a memory and at least one processor, for each of a plurality of time intervals, a likelihood of a particular member of an online social network service performing a particular member user action on a particular message content item during the corresponding time interval; ranking the plurality of time intervals, based on the determined likelihoods corresponding to the plurality of time intervals; identifying a particular time interval from among the plurality of time intervals that is associated with a highest ranking; and classifying the particular time interval as an optimum personalized message delivery time for the particular member; wherein the determining comprises: accessing, via one or more data sources, data including email content data describing a particular email content item and member email interaction data describing the particular member's interactions with various email content; encoding the data accessed from the external data sources into one or more feature vectors, and assembling the one or more feature vectors to thereby generate an assembled feature vector; and performing prediction modeling, based on the assembled feature vector and a trained prediction model, to predict the likelihood of the particular member performing the particular user action on the particular email content item.