US Patent 11386356 Method of training a learning system to classify interfaces

A first path associated with a first interface of a first provider is determined, with the first interface being of a first type. A second path associated with a second interface of the first provider is determined, with the second interface being of a second type. The first path is determined to be a mismatch to the second path. A third path corresponding to an additional interface of the first type is obtained from a second provider. A dictionary is generated based on the third path matching the first path. Vectors corresponding to the first and second types are generated based on the dictionary. A machine learning algorithm is trained based on the vectors with the first and second types as a ground truth values.