Inverse Transform Sampling for Bibliometric Literature Analysis

Abstract

Scientific literature is prosperously evolving, exhibiting exponential growth in the last decades. For a wide range of scientific thematic areas, it is hard or even impossible for individual researchers to analyze in detail the available published works. For this purpose, we utilize a robust multidimensional scaling procedure, to construct the bibliometric maps of the literature, for keywords, authors and references. Particularly, we propose a generic machine learning algorithm for multidimensional scaling, and describe the algorithmic procedure for the generation of the bibliometric maps.