A Research On Article Level Metrics In Research Communities

Since the21st century, every research field is rapidly developing, and the number of scientific literature and research communities are growing. In this situation, it is impossible for every researcher to read every paper related to his research topic. In this situation, how to effectively choose papers to read is important for researchers. At the same time, with the development of Web2.0, Web2.0carries not only new social tools for scientific communication but new article level metrics. How to effectively use these tools to assess the quality of papers also becomes a hot topic.Although some researchers has suggested the metric of social bookmark is effective in assessing for the recently published papers, no one proves whether this metric has more rapid feedback than citation or not and analyses the character of metric itself. In this paper, we conduct quantitative analysis to analyze these two problems and correlation among metrics. The results not only show the social bookmark really has a more rapid feedback than citation but show there is a high correlation among citation, usage and social bookmark. So how to systematically make use of the merits of social bookmark and citation is important.In this paper, we propose article level metric in research communities. The new method of assessing papers not only makes use of the merits of social bookmark and citation, but also takes into account the impact of authors and journals on the quality of recently published papers. Moreover, the new metric makes use of the articles’ listing in the research community. The new metrics provides an easy and effective ways for assessing articles. Experimental results on a sample of papers in the research community of H-index show the efficiency and robust of our new article level metric in assessing the quality of papers and helping researchers to decide which papers to read.Finally, we design the H-index search system in order to make use of the method proposed in this paper. The functions of this system includes searching papers about H-index and show top10researchers, journals and papers in the current year. This system is useful to help researchers to know the latest advances about the research topic of H-index.