Graph-Based Representations in Pattern Recognition: 13th IAPR-TC-15 International Workshop, GbRPR 2023, Vietri sul Mare, Italy, September 6–8, 2023, Proceedings (Lecture Notes in Computer Science)

✍ Scribed by Mario Vento (editor), Pasquale Foggia (editor), Donatello Conte (editor), Vincenzo Carletti (editor)

• Publisher: Springer
• Year: 2023
• Tongue: English
• Leaves: 193
• Category: Library

✦ Synopsis

This book constitutes the refereed proceedings of the 13th IAPR-TC-15 International Workshop on Graph-Based Representations in Pattern Recognition, GbRPR 2023, which took place in Vietri sul Mare, Italy, in September 2023.
The 16 full papers included in this book were carefully reviewed and selected from 18 submissions. They were organized in topical sections on graph kernels and graph algorithms; graph neural networks; and graph-based representations and applications.

✦ Table of Contents

Preface
Organization
Abstracts of Invited Talks
From LBP on Graphs to Slopes in Images
Face to Face: Graphs and Biotechnology
Contents
Graph Kernels and Graph Algorithms
Quadratic Kernel Learning for Interpolation Kernel Machine Based Graph Classification
1 Introduction
2 Interpolating Classifiers
2.1 Interpolation Kernel Machines
3 Multiple Kernel Learning for Interpolation Kernel Machines
3.1 Construction of Combined Kernels
3.2 General Scheme of MKL for Interpolation Kernel Machines
3.3 Dealing with Indefinite Kernels
4 Experimental Results
5 Conclusion
References
Minimum Spanning Set Selection in Graph Kernels
1 Introduction
2 Support Vector Machines and Graph Kernels
2.1 Graph Kernels
3 Kernel Spanning Set Selection via RRQR
4 Experiments and Discussion
5 Conclusion
References
Graph-Based vs. Vector-Based Classification: A Fair Comparison
1 Introduction
2 Research Context
2.1 Graph Classification
2.2 Classification Methods Comparison
3 Experimental Setup
3.1 Datasets
3.2 Experimental Setup
4 Experimental Evaluation
4.1 Graph Classification
4.2 Dataset Selection
5 Conclusions
References
A Practical Algorithm for Max-Norm Optimal Binary Labeling of Graphs
1 Introduction
2 Background and Motivation
3 Preliminaries
3.1 Boolean 2-Satisfiability
3.2 The Malmberg-Ciesielski Algorithm
4 Proposed Algorithm
5 Evaluation
6 Conclusions
References
An Efficient Entropy-Based Graph Kernel
1 Introduction
2 Related Work and Motivation
3 Von Neumann Entropy Based Graph Kernel
4 Evaluation
5 Conclusion
References
Graph Neural Networks
GNN-DES: A New End-to-End Dynamic Ensemble Selection Method Based on Multi-label Graph Neural Network
1 Introduction
2 Graph Neural Network Dynamic Ensemble Selection Technique
3 Experiments
4 Conclusion
References
C2N-ABDP: Cluster-to-Node Attention-Based Differentiable Pooling
1 Introduction
2 Background
2.1 Graph Convolutional Networks
2.2 The Attention Mechanism
2.3 Graph Pooling
2.4 Singular Value Decomposition
3 The Proposed Pooling Method
4 Experiments
4.1 Results and Analysis
5 Conclusion
References
Splitting Structural and Semantic Knowledge in Graph Autoencoders for Graph Regression
1 Introduction
2 Related Work
2.1 Autoencoders
2.2 Graph Autoencoders
3 Proposed Approach: ReGenGraph
3.1 The Learning Process
4 Experimental Validation
4.1 Database
4.2 Architecture Configuration
4.3 Binding Energy Prediction
5 Conclusions
References
Graph Normalizing Flows to Pre-image Free Machine Learning for Regression
1 Introduction
2 Normalizing Flow Preliminaries
3 Proposed Approach
3.1 Regression-Based NF
3.2 Operating in Z
3.3 Pre-imaging
4 Experiments
5 Conclusion
References
Matching-Graphs for Building Classification Ensembles
1 Introduction
2 Building an Ensemble with Matching-Graphs
2.1 Matching-Graphs
2.2 Bagging with Matching-Graphs
3 Experimental Evaluation
3.1 Data Sets and Experimental Setup
3.2 Reference Systems
3.3 Test Results and Discussion
4 Conclusion and Future Work
References
Maximal Independent Sets for Pooling in Graph Neural Networks
1 Introduction
2 Maximal Independent Sets and Graph Poolings
2.1 Maximal Independent Set (MIS) and Meer's Algorithm
2.2 Maximal Independent Sets for Graph Pooling
3 Experiments
4 Conclusion
References
Graph-Based Representations and Applications
Detecting Abnormal Communication Patterns in IoT Networks Using Graph Neural Networks
1 Introduction
2 Representing Network Traffic as a Graph
2.1 Similarity Graphs
2.2 Traffic Trajectory Graphs
2.3 Extended Traffic Dispersion Graphs
3 Graph Neural Networks for Anomaly Detection
3.1 DOMINANT
3.2 OCGNN
3.3 CONAD
4 Experiments
4.1 Datasets
4.2 Graph Neural Network Training
4.3 Results
5 Conclusions
References
Cell Segmentation of in situ Transcriptomics Data Using Signed Graph Partitioning
1 Introduction
1.1 Tools for Analyzing IST Data
1.2 Contribution
2 Methodology
2.1 Compositional Features
2.2 Graph Construction and Partitioning
2.3 Pre and Post Processing
2.4 Visualization
3 Experiments
3.1 osmFISH
3.2 In Situ Sequencing
4 Discussion
References
Graph-Based Representation for Multi-image Super-Resolution
1 Introduction
1.1 Related Work
1.2 Contribution
2 Proposed Method
2.1 Data Representation
2.2 Graph Neural Network
3 Experiments
4 Conclusions
References
Reducing the Computational Complexity of the Eccentricity Transform of a Tree
1 Introduction
2 Definitions
3 Tree Structure
3.1 Line Segment
3.2 Branching Point
3.3 Tree
4 Shape
4.1 3D Curve of a Shape
4.2 Smooth Shapes Without Holes
5 Simulation and Result
6 Conclusion
References
Graph-Based Deep Learning on the Swiss River Network
1 Introduction
2 Related Work
2.1 Air2Stream
2.2 LSTM on Water Data
3 The Swiss Water Body Graph
3.1 Construction of the Graph
3.2 Proposed Water Challenges
4 Proposed Method and Experimental Evaluation
4.1 Experimental Setup and Reference Models
4.2 The Novel Graph-Based Model
4.3 Test Results
5 Conclusion and Future Work
References
Author Index