International Journal Impact Factor Modeling Based on Gene Expression Programming

Academic periodical is able to manifest scientific competitiveness of one country, and also is an important manifestation form of scientific productivity. Therefore, the research on international journal impact factor modeling has certain practical significance. Based on the gene expression programming, this paper establishes nonlinear model between journal impact factor and related indexes. The model utilizes immediacy index, total cites, articles, cited half-life, issues per year, self cites and average references as the inputs, and takes impact factor as the output. This paper mainly chooses the academic journals of China in 2008 as experimental data that are embodied by Journal Citation Reports. The experimental results show that impact factor is positively related to immediacy index, total cites and average references, and is negatively related to articles and cited half-life, and nearly has nothing to do with issues per year and self cites.