Bipartite Yule Processes in Collections of Journal Papers

Abstract

Collections of journal papers, often referred to as 'citation networks', can be modeled as a collection of coupled bipartite networks which tend to exhibit linear growth and preferential attachment as papers are added to the collection. Assuming primary nodes in the first partition and secondary nodes in the second partition, the basic bipartite Yule process assumes that as each primary node is added to the network, it links to multiple secondary nodes, and with probability, $α$, each new link may connect to a newly appearing secondary node. The number of links from a new primary node follows some distribution that is a characteristic of the specific network. Links to existing secondary nodes follow a preferential attachment rule. With modifications to adapt to specific networks, bipartite Yule processes simulate networks that can be validated against actual networks using a wide variety of network metrics. The application of bipartite Yule processes to the simulation of paper-reference networks and paper-author networks is demonstrated and simulation results are shown to mimic networks from actual collections of papers across several network metrics.