Research On Geographic Dissemination Of Papers Based On Readers' Location Data

With the widespread application of big data,the usage data of papers has gradually entered the research field of information measurement.This paper selects the reading data of billions of SCI papers published by Frontiers as the research object,and conducts the analysis of the characteristics of geographic dissemination of scientific papers combining statistical analysis,correlation analysis,and visual analysis.Firstly,Spearman correlation analysis is conducted to explore the relationship between publication quantity and reading quantity in different countries based on SPSS software.And it was found that there was a strong positive correlation between the distribution areas of the authors and that of the readers,which means where there is high publication quantity,there is high reading quantity,and the opposite is also correct.To be further,five cities with the highest publications are selected as research objects to study the geographical diffusion source,diffusion center,and diffusion radius.And the results show that both the geographic diffusion source and center are localized,and both the two are consistent with the author's city or country.In addition,the diffusive speed of radius changes from fast to slow in short-term,and the radius will reach a basically stable state in long term.Finally,we visualize the short-term and long-term distribution of the cities with most readers,and the conclusion shows that the cumulative number of reading in the authors' cities is much higher than that of others both in short term and long term after the publication of papers,however,the cumulative number of reading in areas inconsistent with the countries of authors is relatively lower than that of others.Moreover,the accumulated reading times of the cities belonging to the countries of authors but inconsistent with the cities of authors are relatively high in the short term,and low in the long term.