A Bibliometric Framework For Identifying Landmark Papers Within A Filed Based On Citation Network

Academic inheritance is an objective pattern of academic relationship in the certain academic field.In basic research of natural science,the academic research achievements are mainly represented by academic papers and academic inheritance is mainly represented by academic chain.The identification of Landmark Papers within a filed which are the nodes of academic chain can promote the quantitative study of the history of science and provide a new perspective for the evaluation of academic influence of papers.The citation relationship between papers has become a norm.The relationship between the current research results and the research foundation of the predecessors and the progress of the later generations can be established by the citation mechanism,and the method of identifying the Landmark Papers based on it has become the mainstream.After reviewing the existing identification methods,this article constructed a bibliometric framework which is based on the citation network objectively formed by the paper citation relationship and gives consideration to the impact,vitality and acceptance of Landmark Papers.Two subject areas of different disciplines were selected for empirical analysis,and the result was evaluated by referring to the "accuracy" and "recall" in the field of information retrieval.It is found that the framework constructed in this paper can generally identify Landmark Papers.Citation strength and citation network can complement each other and identify the papers as comprehensive as possible,and the citation duration and the citation content can improve the accuracy under the condition of ensuring a certain recall.From the comparative analysis of two cases,it is found that the identification framework constructed in this paper is more suitable for application-oriented research than basic scientific research.Finally,this paper took an example to analyze the characteristics of the citation content of Landmark Papers,and proposed suggestions for the improvement of the identification method.