Learning the Sub-optimal Graph Edit Distance Edit Costs Based on an Embedded Model

摘要

Graph edit distance has become an important tool in structural pattern recognition since it allows us to measure the dissimilarity of attributed graphs. One of its main constraints is that it requires an adequate definition of edit costs, which eventually determines which graphs are considered similar. These edit costs are usually defined as concrete functions or constants in a manual fashion and little effort has been done to learn them. The present paper proposes a framework to define these edit costs automatically. Moreover, we concretise this framework in two different models based on neural networks and probability density functions.