UsefulLinks
Computer Science
Artificial Intelligence
Deep Learning
Graph Neural Networks
1. Foundations for Understanding GNNs
2. Core Concepts of Graph Neural Networks
3. Foundational GNN Architectures
4. Advanced GNN Architectures
5. GNN Training and Optimization
6. GNN Applications and Tasks
7. Evaluation and Benchmarking
8. Advanced Topics and Research Frontiers
9. Implementation and Practical Considerations
7.
Evaluation and Benchmarking
7.1.
Evaluation Metrics
7.1.1.
Classification Metrics
7.1.1.1.
Accuracy
7.1.1.2.
Precision and Recall
7.1.1.3.
F1-Score
7.1.1.4.
Macro and Micro Averaging
7.1.1.5.
Area Under Curve (AUC)
7.1.2.
Regression Metrics
7.1.2.1.
Mean Absolute Error
7.1.2.2.
Root Mean Square Error
7.1.2.3.
R-squared
7.1.2.4.
Mean Absolute Percentage Error
7.1.3.
Ranking Metrics
7.1.3.1.
Precision at K
7.1.3.2.
Recall at K
7.1.3.3.
Normalized Discounted Cumulative Gain
7.1.3.4.
Mean Reciprocal Rank
7.1.4.
Link Prediction Metrics
7.1.4.1.
AUC-ROC
7.1.4.2.
AUC-PR
7.1.4.3.
Hits at K
7.1.4.4.
Mean Rank
7.2.
Experimental Design
7.2.1.
Data Splitting Strategies
7.2.1.1.
Random Splits
7.2.1.2.
Temporal Splits
7.2.1.3.
Structural Splits
7.2.2.
Cross-validation
7.2.2.1.
K-fold Cross-validation
7.2.2.2.
Leave-one-out
7.2.2.3.
Stratified Sampling
7.2.3.
Statistical Significance Testing
7.2.3.1.
Paired t-tests
7.2.3.2.
Wilcoxon Signed-rank Test
7.2.3.3.
Multiple Comparison Correction
7.3.
Benchmark Datasets
7.3.1.
Node Classification Benchmarks
7.3.1.1.
Citation Networks
7.3.1.2.
Social Networks
7.3.1.3.
Biological Networks
7.3.2.
Graph Classification Benchmarks
7.3.2.1.
Molecular Datasets
7.3.2.2.
Social Network Datasets
7.3.2.3.
Synthetic Datasets
7.3.3.
Link Prediction Benchmarks
7.3.3.1.
Knowledge Graphs
7.3.3.2.
Social Networks
7.3.3.3.
Biological Networks
7.4.
Reproducibility and Fair Comparison
7.4.1.
Hyperparameter Tuning
7.4.2.
Model Selection
7.4.3.
Reporting Standards
7.4.4.
Code and Data Availability
Previous
6. GNN Applications and Tasks
Go to top
Next
8. Advanced Topics and Research Frontiers