A non-linear index to evaluate a journal's scientific impact

Abstract

The purpose of this study is to define a bibliometric indicator of the scientific impact of a journal, which combines objectivity with the ability to bridge many different bibliometric factors and in particular the side factors presented along with celebrated ISI impact factor. The particular goal is to determine a standard threshold value in which an independent self-organizing system will decide the correlation between this value and the impact factor of a journal. We name this factor "Cited Distance Factor (CDF)" and it is extracted via a well-fitted, recurrent Elman neural network. For a case study of this implementation we used a dataset of all journals of cell biology, ranking them according to the impact factor from the Web of Science Database and then comparing the rank according to the cited distance. For clarity reasons we also compare the cited distance factor with already known measures and especially with the recently introduced eigenfactor of the institute of scientific information (ISI).