Fast and Flexible Ranking on Bipartite Networks with Python and R
Bipartite (two-mode) networks are ubiquitous. Common examples include networks of collaboration between scientists and their shared papers and networks of affiliation between corporate directors and board members.
Bipartite networks are commonly reduced to unipartite networks for further analysis, such as calculating node centrality (e.g. PageRank). However, one-mode projections often destroy important structural information and can lead to imprecise network measurements. Moreover, there are numerous ways to obtain unipartite networks from a bipartite network, each of which has different characteristics and idiosyncrasies. To overcome the issues of one-mode projection, we develop BiRank, a Python and R package that calculates node centrality on bipartite networks directly.
In this workshop, Kaicheng Yang (PhD student, Complex Networks and Systems, IUB) will introduce the algorithms implemented in the BiRank package and demonstrate how to use them. He’ll then provide a real-world research question and show how to use this package to analyze the data and interpret the results.
Prerequisites: A familiarity with basic network science concepts and programming languages such as Python and/or R.
This workshop will be held on-line, with a limited number of seats available in-person. Both attendance options require registration. Click here to register.
Return to the workshop series home page.