Modelling multi-relational data with special attention to the average number of collaborators as a variable in informetric distributions

Abstract

We prove a theorem about the invariance of the Lotka function under a transformation which maps numbers of authors as variables to number of collaborators as variables. Moreover, we describe a general approach to incorporate the average number of collaborators in a production function, even when the conditions of the theorem are not satisfied. This leads to a general framework to model multi-relational data. Several cases are described where this general model can be applied.