Using outcome-targeted gap predictions to identify a digital resource

A computer program product comprising one or more non-transitory computer-readable storage media comprising instructions which, when executed by one or more processors, cause:

• extracting, from an online connection network, digital data comprising target profile data and current profile data;
• wherein the target profile data is associated with an online submission process;
• wherein the current profile data is associated with a node of the online connection network;
• generating, using a machine learning model, a boundary model based on a portion of the digital data;
• predicting, using a prediction model, attribute data that is associated with a positive outcome based on the boundary model and the digital data;
• classifying, by the machine learning model, the target profile data on a side of a boundary line of the boundary model that is associated with positive outcomes;
• computing gap data as a measure of a distance between the current profile data and the side of a boundary line of the boundary model that is associated with positive outcomes;
• inputting the gap data into an active learning process;
• identifying, by the active learning process, a sub-portion of attribute data that is in the target profile data but is not in the current profile data;
• determining that the attribute data added to the current profile data, minimizes the distance between the current profile data and the side of the boundary line of the boundary model that is associated with the positive outcomes;
• outputting the attribute data to a downstream process or an automated digital assistant or the online connection network or through an online learning system;
• and
• associating, based on the sub-portion of attribute data, a particular segment of video content with the node through the downstream process or the automated digital assistant or the online connection network or the online learning system.