Interdisciplinary Research On Public Administration Based On Rao-Stirling Diversity Measure

With the increasingly complex problems in the public administration sphere and the increasingly diversified objects involved in public administration affairs,it is urgent to cultivate high-level compound talents in the field of public administration in order to cope with more and more difficult high-level public administration affairs.In the early years,the way of personnel training that relied too much on the professional knowledge of humanities and social sciences but ignored the professional knowledge of science and engineering became more and more rigid,which made it more and more difficult to meet the needs of today's society.It was this reflection on the demand for talents and the mode of personnel training that led to the demand-oriented interdisciplinary development of public administration sphere.As the traditional paradigm and model of public administration need to be upgraded and renewed urgently,the disciplinary barriers between public administration disciplines and various disciplines have been gradually broken,and the integration of different disciplines has begun.Public administration disciplines are constantly interdisciplinary with other disciplines,and their interdisciplinary nature has been greatly improved.Faced with the increasing interdisciplinary nature of public administration disciplines,how to evaluate the interdisciplinary nature of current public administration disciplines,which disciplines have made important contributions to the development of public administration disciplines and which disciplines may become powerful promoters of the future development of public administration have become a new and attractive research issue.Interdisciplinary research mainly focuses on academic papers,and reflects the degree of interdisciplinary nature of the disciplines to which the papers belong by evaluating the degree of interdisciplinary nature of the papers.However,the current interdisciplinary research is in its infancy,and there are many shortcomings in the existing interdisciplinary evaluation methods.Rao-Stirling diversity measure is one of the most widely used interdisciplinary evaluation methods,which contains three important attributes of interdisciplinary: variety,balance and diversity.However,the traditional Rao-Stirling diversity measure has two shortcomings: one is that it does not consider the quality factors in interdisciplinary research,that is,it does not consider the impact of the quality of the paper itself on its interdisciplinary degree.If a target paper is cited by a high-quality paper and a low-quality paper at the same time,the high-quality paper will be inferior to the target paper.Quantitative papers have higher interdisciplinary contribution,that is,target papers cited by high-quality papers will be more interdisciplinary than those cited by low-quality papers;secondly,the particularity of author attributes in interdisciplinary research is not considered.In interdisciplinary research,author-link relationship is a special kind of relationship,among papers and papers,journals and periodicals,and between disciplines and disciplines,all can rely on citation relationship,but only between authors and authors rely on cooperative relationship.The traditional Rao-Stirling diversity measure only includes citation relationship,but does not include cooperative relationship.A Rao-Stirling diversity method involving collaborative relationships is proposed to assess the interdisciplinary nature of authors.This paper takes the academic papers of public administration as the research object,and makes a thorough study of the Rao-Stirling diversity measure in interdisciplinary research based on considering the quality attribute and the author attribute respectively.In this paper,a Rao-Stirling diversity measure based on quality attributes is proposed to solve the problem that the traditional Rao-Stirling diversity measure does not consider the influence of the paper's own quality factors on its interdisciplinary degree.Secondly,a Rao-Stirling diversity measure based on authors' attributes is proposed to solve the problem that the traditional Rao-Stirling diversity measure does not consider the interdisciplinary particularity of authors due to the particularity of authors' links.Finally,according to the characteristics of the two improved Rao-Stirling diversity measures mentioned above,the corresponding paper data sets are selected to solve their interdisciplinary degree,and the rationality and superiority of the two improved Rao-Stirling diversity measures are verified.The main contributions of this paper can be concluded as follows:1.In order to consider the influence of quality factors on the interdisciplinary degree of the paper,P-Rank algorithm is introduced to solve the quality value of the paper itself.The P-Rank algorithm includes one isomorphic network and two heterogeneous networks.The isomorphic network is used to describe the citation relationship between papers.The two heterogeneous networks are used to describe the paper-author relationship and the paper-journal relationship,respectively.Based on the principle of quality transfer among three networks,the quality values of the target papers are obtained.Then,each target paper is assigned according to its quality.The traditional Rao-Stirling diversity method is introduced to optimize the richness attributes of Rao-Stirling diversity measure,and the Rao-Stirling diversity measure based on quality attributes is obtained.2.In order to consider the particularity of author links in interdisciplinary research,the author cooperative network similarity algorithm is introduced to replace the reference relationship in the traditional Rao-Stirling diversity measure with the author cooperative relationship.Firstly,the author's cooperative network similarity algorithm is used to solve the similarity between disciplines with citation relationship and form the similarity matrix between disciplines.Then,a new interdisciplinary similarity matrix is used to replace the interdisciplinary similarity matrix based on cosine similarity algorithm in the traditional Rao-Stirling diversity measure.In addition,the existing similarity algorithm of author cooperative network is an interdisciplinary similarity algorithm based on author centrality.Considering the advantage of author centrality algorithm in evaluating interdisciplinary degree,this paper replaces author centrality algorithm in author centrality algorithm and obtains author attribute-based Rao-Stirling diversity measure.