Toggle navigation sidebar
Toggle in-page Table of Contents
Hands-on Network Machine Learning with Scikit-Learn and Graspologic
Hands-on Network Machine Learning with Scikit-Learn and Graspologic
Introduction
Preface
Terminology
Foundations
1. The Network Machine Learning Landscape
1.1. What is network machine learning?
1.2. Why do we study networks?
1.3. Types of Network Machine Learning Problems
1.4. Challenges of Network Machine Learning
2. End-to-end Biology Network Machine Learning Project
2.1. Look at the big picture
2.2. Get the Data
2.3. Prepare the Data for Network Algorithms
2.4. Select and Train a Model
2.5. Fine-Tune your Model
2.6. Discover and Visualize the Data to Gain Insights
Representations
3. Properties of Networks as a Statistical Object
3.1. Matrix Representations Of Networks
3.2. Properties of Networks
3.3. Representations of Networks
3.4. Regularization
4. Why Use Statistical Models?
4.1. Erdös-Rényi (ER) Random Networks
4.2. Stochastic Block Models (SBM)
4.3. Random Dot Product Graphs (RDPG)
4.4. Inhomogeneous Erdos Renyi (IER) Random Network Model
4.5. Degree-Corrected Stochastic Block Model (DCSBM)
4.6. Structured Independent Edge Model (SIEM)
4.7. Multiple Network Models
4.8. Network Models with Network Covariates
5. Learning Network Representations
5.1. Estimating Parameters in Network Models via MLE
5.2. Why embed networks?
5.3. Spectral embedding methods
5.4. Multiple-Network Representation Learning
5.5. Joint Representation Learning
Applications
6. Applications When You Have One Network
6.1. Community Detection
6.2. Testing for Differences between Groups of Edges
6.3. Model Selection
6.4. Single-Network Vertex Nomination
6.5. Out-of-sample Embedding
7. Applications for Two Networks
7.1. Latent Two-Sample Hypothesis Testing
7.2. Two-sample hypothesis testing in SBMs
7.3. Graph Matching
7.4. Vertex Nomination For Two Networks
8. Applications for Many Networks
8.1. Anomaly Detection For Timeseries of Networks
8.2. Testing for Significant Edges
8.3. Testing for Significant Vertices
Next Steps
9. Where do we go from here?
9.1. Random walk and diffusion-based methods
9.2. Graph Neural Networks
9.3. Network Sparsity
Appendix
10. Representations (Extended)
10.1. Alternative Network Representations
11. Network Model Theory
11.2. Background
11.3. Foundation
11.4. Erdös-Rényi (ER) Random Networks
11.5. Stochastic Block Models
11.6. RDPGs and more general network models
12. Learning Representations Theory
12.1. Maximum Likelihood Estimate Theory
12.2. Spectral Method Theory
13. Applications (Extended)
13.1. Hypothesis Testing with coin flips
13.2. Unsupervised learning
13.3. Bayes Plugin Classifier
Reference
Graspologic Documentation
repository
open issue
Index
