Knowledge Evolution Path Analysis Of Domestic Blockchain Research

With the general trend of industrial development and new technological innovation,the role of blockchain technology has become increasingly prominent and has become an important battlefield for a new round of competitive development in new technological and industrial upgrading.As a result,the number of scientific research documents on the subject of blockchain has increased dramatically,and more and more scholars have joined the research of the practical application.However,blockchain research is an emerging field with the characteristics of rapid development,relatively plain research content,broad research scope,as well as the complicate related research fields.it is difficult for researchers to review the research direction and track the development of the concept of the domain,so it is important to clarify the domain system of blockchain and grasp the related research context for the future research.This paper has fully studied domestic and foreign research,founding that the knowledge evolution path can show the generation and development of knowledge.It is a powerful tool to clarify the development law of the field(discipline).As one of the important methods of information measurement,Main Path Analysis is a common method of trunk extraction in citation networks and an important means of generating knowledge evolution path,which can be used to describe domain(discipline)evolution path.However,the traditional Main Path Analysis has some limitations,such as generated a single path and the inability to reflect text information.Therefore,in the process of studying the document in the domestic blockchain field,this paper summarizes a knowledge evolution path analysis model that is different from previous studies and is suitable for the development of emerging fields.From the perspective of evolution,text information is introduced into the main path document node to reflect the change of the research topic of blockchain.This model can present literature topics in the knowledge evolution path so that researchers can quickly and accurately grasp the development of the topic.At the same time,the model can help relevant scholars to identify the topic discovery and hotspot changes in this field,and provide a reference for relevant scholars to clarify the development path of the blockchain and track the hotspots of blockchain research.The core content of this paper focuses on two parts: generating the combined main path of domestic blockchain research and analyzing the evolution path of blockchain knowledge based on similar documents.In the chapter of generating the combined main path of the domestic blockchain research,the purpose is to focus on and improve the problems of the single path of the traditional Main Path Analysis and the weak text combination found in the early stage.First,the target documents were obtained from CNKI.And then,the Search Path Count(SPC)algorithm is used to extract and construct the main path in the domestic blockchain document citation network,that is,to obtain the traditional main path.The main path is extracted separately according to the generated cluster tags,and the multiple "blockchain + theme" main paths are connected according to the nodes with large mediation center to form the combined main path.Finally,the results of the new combined main path are compared with those of the traditional main path,and the advantages of the combined main path are analyzed.The combined main path improves the shortcomings of the traditional main path.However,the combined main path has less semantic information,which is not conducive for researchers to quickly obtain the development trend and hot information of blockchain research.Therefore,based on the obtained combined main path,this paper explores the method of supplementing the main path analysis semantic information,and generates a blockchain knowledge evolution path analysis based on similar documents.In this part of the content,first we use the cosine similarity formula in the Vector Space Model to calculate the similarity value between the non-main path documents and the obtained main path documents.Due to the large amount of data in the non-main path documents,the similarity threshold is difficult to define.So this paper introduces high-cited indicators to select the similar documents,and adds the selected non-main path documents to the original combination main path nodes to enrich the content of the original path nodes.Secondly,we extract the subject words of each node document group through the LDA model,and add the subject words to the original combined main path to form an information-rich domestic blockchain research knowledge evolution path to fully present the development trend of the blockchain.