Titlebook: Structural Pattern Recognition with Graph Edit Distance; Approximation Algori Kaspar Riesen Book 2015 Springer International Publishing Swi

一致性

Learning Exact Graph Edit Distanceres—which numerically characterizes individual node edit operations—is defined and extracted from the underlying graphs. These features are in turn used for the development of a classification model for node edit operations.

NAUT

尽管

Introduction and Basic Concepts structural pattern recognition. In particular, we discuss the advantages and drawbacks of both approaches. Eventually, graph-based pattern representation is formally introduced and complemented by a list of applications where graphs are actually employed. The remaining parts of this chapter are the

Panacea

使隔离

无孔

initiate

Learning Exact Graph Edit Distance In the present chapter, two additional approaches are pursued for reducing the approximation error. First, in Sect. . we introduce a method that aims at estimating the exact edit distance . based on both distance bounds . and . derived from BP-GED. This method is based on regression analysis. Secon

FILTH

Speeding Up Bipartite Graph Edit Distancepeeding up the second step of BP-GED, viz., the assignment of local substructures in the LSAP. In BP-GED a state-of-the-art algorithm with cubic time complexity is employed in order to optimally solve the underlying LSAP. In this chapter, we aim at replacing this optimal algorithm with suboptimal gr

interlude

Conclusions and Future Workand evaluated. This last chapter first summarizes and comments the main contributions of the present book. Eventually, the main findings—that have been achieved by means of the experimental evaluations carried out in various chapters—are brought together. Finally, we show several rewarding options f

轻推

Introduction and Basic Conceptsn dedicated to formal introductions of diverse graph matching definitions. We particularly delve into the difference between exact and inexact graph matching. Last but not least, we give a brief survey of existing graph matching methodologies that somehow differ from the approach that is actually pursued in the present book.