Research On Evolutionary Analysis Of Cross Disciplinary Themes Based On Direct Citation Content

In the context of today’s big science era,science and technology are changing with each passing day,and the scientific research problems facing mankind are becoming more and more diversified and complex.It is difficult to fundamentally solve them by relying on a single discipline.Therefore,interdisciplinary knowledge exchange and integration are increasingly frequent,and the barrier between disciplines is broken.As the most core mapping of discipline integration,the effective identification and analysis of its evolution trend is not only conducive to breaking the traditional discipline boundaries,mining new growth points of disciplines,but also helps researchers sort out the new knowledge association system established by basic disciplines,and further promote the coordination,development,integration,intersection and innovation of disciplines.However,there are few studies on the evolution of interdisciplinary subjects,and most of them use a single citation analysis or text analysis.The detection method that organically integrates the two needs to be further explored.Therefore,this study constructs an interdisciplinary subject evolution model combined with citation content analysis.On the basis of finding highly interdisciplinary literature by using direct citation,it further uses text mining technology to identify the subject of the interdisciplinary literature,determine the development stage of the subject,grasp the development trend of the interdisciplinary field in the dynamic change,and take the knowledge atlas of the common research field of library and information science and computer science as an example,Verify the feasibility of the model.Specifically,this study is mainly divided into three parts: First,based on the review of the current research situation at home and abroad,the basic issues of the construction of the technical route of interdisciplinary evolution analysis in discipline fields,including definitions,methods,and measurement indicators,are described.Second,use the "literature direct citation",combine with the "LDA theme model",introduce the time factor,comprehensively design the technical route of interdisciplinary evolution analysis,deeply mine the internal knowledge of interdisciplinary literature in dynamic changes,and scientifically determine the development stage(potential,development,maturity,and decline)of the interdisciplinary fields.Thirdly,the feasibility of the model is verified by taking the cross literature of library and information science and computer science in the field of knowledge atlas in the Web of Science database as an example.The result shows that the technical route of interdisciplinary evolution analysis based on direct citation content is feasible.Direct citation of documents does not rely on third parties,directly reflects the inheritance relationship between documents,and has certain natural advantages in detecting knowledge exchange and transfer;The LDA topic model has complete theory,clear structure,and good generalization ability.It can preserve the internal relationship of the subject to the greatest extent in the topic mining.The paper organically unifies these two aspects and gives full play to their respective advantages.Compared with the traditional single analysis method,it is more scientific,understandable,stable and convenient.