Tutorials#
The narrative tutorials explain why a method behaves as it does, with figures computed when the documentation is built. The how-to tutorials show the calls, method by method.
Start here#
New to backbones? Read this worked case study first: it takes two weighted networks from a tangled drawing to a reported backbone, explaining why each step is needed (scoring and filtering, choosing alpha and corrections, comparing methods, validating, reporting), with figures computed when the docs are built.
By network type#
Narrative tutorials for each kind of network:
Short how-to pages for each kind of network:
- Statistical Backbone Extraction
- Proximity-Based Edge Scoring
- Structural Backbone Methods
- Bipartite Projection Backbones
- What is a bipartite projection backbone?
- Setup
- Weighted projections
- SDSM: Stochastic Degree Sequence Model
- FDSM: Fixed Degree Sequence Model
- Additional null models
- Comparing SDSM and FDSM
- Partition selection note
- Incidence matrices and large data
- Signed backbones
- Edge constraints (SDSM-EC)
- References
- Hypergraph Backbones
- Unweighted Graph Sparsification
- Temporal Backbones
Going further#
Narrative deep dives, and comparing, validating, and benchmarking backbones.