A system for improving distributed network data flow efficiency by using a machine learning model to engage a reduced set of unknown objects is disclosed. The system includes at least one processor, a communication interface communicatively coupled to the at least one processor, and a memory device storing executable code that, when executed causes the at least one processor to receive interaction data corresponding to interactions between known objects and unknown objects, develop an interaction map, develop a profile for each unknown object, process each unknown object using a machine learning algorithm to generate a reduced set of unknown objects most likely to reciprocate and identify properties of each unknown object, and trigger a communication to each of the reduced set of unknown objects.