Structural Pattern Recognition With Graph Edit Distance

This unique text/reference presents a thorough introduction to the field of structural pattern recognition, with a particular focus on graph edit distance (GED), one of the most flexible graph distance models available. The book also provides a detailed review of a diverse selection of novel methods related to GED, and concludes by suggesting possible avenues for future research. Topics and features: Formally introduces the concept of GED, and highlights the basic properties of this graph matching paradigm Describes a reformulation of GED to a quadratic assignment problem Illustrates how the quadratic assignment problem of GED can be reduced to a linear sum assignment problem Reviews strategies for reducing both the overestimation of the true edit distance and the matching time in the approximation framework Examines the improvement demonstrated by the described algorithmic framework with respect to the distance accuracy and the matching time Includes appendices listing the datasets employed for the experimental evaluations discussed in the book Researchers and graduate students interested in the field of structural pattern recognition will find this focused work to be an essential reference on the latest developments in GED. Dr. Kaspar Riesen is a university lecturer of computer science in the Institute for Information Systems at the University of Applied Sciences and Arts Northwestern Switzerland, Olten, Switzerland.

In graph-based structural pattern recognition, the idea is to transform patterns into graphs and perform the analysis and recognition of patterns in the graph domain OCo commonly referred to as graph matching. A large number of methods for graph matching have been proposed. Graph edit distance, for instance, defines the dissimilarity of two graphs by the amount of distortion that is needed to transform one graph into the other and is considered one of the most flexible methods for error-tolerant graph matching.This book focuses on graph kernel functions that are highly tolerant towards structural errors. The basic idea is to incorporate concepts from graph edit distance into kernel functions, thus combining the flexibility of edit distance-based graph matching with the power of kernel machines for pattern recognition. The authors introduce a collection of novel graph kernels related to edit distance, including diffusion kernels, convolution kernels, and random walk kernels. From an experimental evaluation of a semi-artificial line drawing data set and four real-world data sets consisting of pictures, microscopic images, fingerprints, and molecules, the authors demonstrate that some of the kernel functions in conjunction with support vector machines significantly outperform traditional edit distance-based nearest-neighbor classifiers, both in terms of classification accuracy and running time."

This unique text/reference presents a thorough introduction to the field of structural pattern recognition, with a particular focus on graph edit distance (GED). The book also provides a detailed review of a diverse selection of novel methods related to GED, and concludes by suggesting possible avenues for future research. Topics and features: formally introduces the concept of GED, and highlights the basic properties of this graph matching paradigm; describes a reformulation of GED to a quadratic assignment problem; illustrates how the quadratic assignment problem of GED can be reduced to a linear sum assignment problem; reviews strategies for reducing both the overestimation of the true edit distance and the matching time in the approximation framework; examines the improvement demonstrated by the described algorithmic framework with respect to the distance accuracy and the matching time; includes appendices listing the datasets employed for the experimental evaluations discussed in the book.

The refereed proceedings of the 4th IAPR International Workshop on Graph-Based Representation in Pattern Recognition, GbRPR 2003, held in York, UK in June/July 2003. The 23 revised full papers presented were carefully reviewed and selected for inclusion in the book. The papers are organized in topical sections on data structures and representation, segmentation, graph edit distance, graph matching, matrix methods, and graph clustering.

This book constitutes the refereed proceedings of the 8th IAPR-TC-15 International Workshop on Graph-Based Representations in Pattern Recognition, GbRPR 2011, held in Münster, Germany, in May 2011. The 34 revised full papers presented were carefully reviewed and selected from numerous submissions. The papers are organized in topical sections on graph-based representation and characterization, graph matching, classification, and querying, graph-based learning, graph-based segmentation, and applications.

This book constitutes the proceedings of the Joint IAPR International Workshop on Structural, Syntactic, and Statistical Pattern Recognition, S+SSPR 2022, held in Montreal, QC, Canada, in August 2022. The 30 papers together with 2 invited talks presented in this volume were carefully reviewed and selected from 50 submissions. The workshops presents papers on topics such as deep learning, processing, computer vision, machine learning and pattern recognition and much more.

This book constitutes the proceedings of the Joint IAPR International Workshop on Structural, Syntactic, and Statistical Pattern Recognition, S+SSPR 2020, held in Padua, Italy, in January 2021. The 35 papers presented in this volume were carefully reviewed and selected from 81 submissions. The accepted papers cover the major topics of current interest in pattern recognition, including classification and clustering, deep learning, structural matching and graph-theoretic methods, and multimedia analysis and understanding.

This book constitutes the proceedings of the Joint IAPR International Workshop on Structural, Syntactic, and Statistical Pattern Recognition, S+SSPR 2014; comprising the International Workshop on Structural and Syntactic Pattern Recognition, SSPR, and the International Workshop on Statistical Techniques in Pattern Recognition, SPR. The total of 25 full papers and 22 poster papers included in this book were carefully reviewed and selected from 78 submissions. They are organized in topical sections named: graph kernels; clustering; graph edit distance; graph models and embedding; discriminant analysis; combining and selecting; joint session; metrics and dissimilarities; applications; partial supervision; and poster session.