An investigation of web resource distribution in the field of information science

Abstract

This study introduces a new methodology to explore Web information distribution. Subject terms extracted from a key journal in the field of information science were employed to conduct Web searches on Google to identify a corpus of Internet domains and associated Web pages to represent the discipline of information science. A Bradford analysis was then applied to the corpus to determine if the scatter of Web pages conformed to a Bradford distribution. The modeling of the collected data to the power law function in LOTKA program indicates a good fit at even 10% significance level. With a binning procedure and least squares fitting, an R square value of 0.987 was obtained. A division of the data according to top level domain category shows different number of domains and domain productivities in different types of domains. Governmental and commercial domains have higher productivity than the educational and organizational domains. However, the difference between governmental and commercial domains and between educational and organizational domains are not significant