implement more measures to aid interpretation

[jboynyc]

For instance Liebig & Rao (2014) to identify influential nodes driving clustering.

[jboynyc]

@BrandonKMLee given your apparent detailed knowledge of the Python ecosystem for network analysis, do you know of any project that implements a clustering coefficient for bipartite networks (or anything comparable)?

(Same question about #24, where I'm collecting ideas for improved backbone extraction of projected networks. I've come across the backbone package for R which implements a few techniques, but nothing comparable in Python land.)

[BradKML]

Firstly, obligatory reference to Centrality #32

Also for Bipartite-specific Metrics, which most are based on HITS or PageRank:

• https://arxiv.org/pdf/0710.0117.pdf
• https://osf.io/preprints/socarxiv/hazvs/download
• https://sci-hub.se/10.1016/j.physa.2008.09.006
• https://sci-hub.se/10.1080/00222500590889730
• https://joss.theoj.org/papers/10.21105/joss.02315.pdf
• https://cs.paperswithcode.com/paper/hellrank-a-hellinger-based-centrality-measure
• https://doi.org/10.1017/dsj.2017.32

Also for NetwrokX, the basic centralities (includes redundancy) and Clustering data exists

• https://networkx.org/documentation/stable/reference/algorithms/bipartite.html#module-networkx.algorithms.bipartite.cluster
• https://networkx.org/documentation/stable/reference/algorithms/bipartite.html#module-networkx.algorithms.bipartite.centrality
• https://networkx.org/documentation/stable/reference/algorithms/bipartite.html#module-networkx.algorithms.bipartite.redundancy

[jboynyc]

Closing for now, as the new release will include bipartite centrality measures and a clustering coefficient. Will likely revisit this in the future.