Systems And Methods Of Modeling Object Networks

摘要

According to one embodiment, a system is provided. The system includes a memory, at least one processor coupled to the memory and an object network modeler component executable by the at least one processor. The memory stores an object network including a plurality of objects, the plurality of objects including a first object, a second object, a third object, and a fourth object. The object network modeler component is configured to implicitly associate, within the object network, the first object with the second object and explicitly associate, within the object network, the third object with the fourth object.

主权项

1. A computing system for identifying objects within an object network that are sufficiently similar to a query, the computing system comprising:
memory comprising executable instructions; and a processor operatively connected to the memory, the processor configured to execute the executable instructions in order to effectuate a method comprising:
obtaining a probe object representative of the query;generating one or more external vectors representative of the probe object;generating one or more internal vectors representative of the probe object based on the one or more external vectors representative of the probe object;generating a fingerprint representative of the probe object based on the one or more internal vectors representative of the probe object;performing at least one of the following comparisons to identify a first set of candidate objects:
comparing the fingerprint representative of the probe object with a plurality of fingerprints representative of a plurality of objects within the object network; andcomparing the one or more internal vectors representative of the probe object with a plurality of internal vectors representative of the plurality of objects within the object network;calculating first respective similarity metrics between the probe object and each of the objects in the first set of candidate objects by comparing the one or more internal vectors representative of the probe object with corresponding internal vectors representative of each object in the first set of candidate objects; andgenerating a second set of candidate objects based on the first set of candidate objects, wherein the second set of candidate objects comprises those objects whose first similarity metrics exceed a first predefined threshold.