Research On The Construction Of Knowledge System In Building Energy Efficiency Based On Bibliographic Data Mining

With the development of information technology,scientific research has been in the data-intensive discovery "fourth paradigm".It is the same in the field of building energy efficiency.In order to reduce building energy use and the environmental crisis,scholars have accumulated a wealth of knowledge research achievements in this field.Literature is the carrier of knowledge,and with the rapid growth of domain literature,domain knowledge information has become more specialized,subdivided and complicated.The knowledge system in the field of building energy efficiency is presented in traditional forms such as textbook system.It is a systematic summary of the basic energy efficiency knowledge and technical principles of various functional systems in buildings.However,under the rapid development of the field,the knowledge presented by existing systems is mature,but the update speed of frontier hotspot knowledge is slow,and the connotation relationship between hotspot knowledge is not clear.Therefore,how to extract the frontier hotspot knowledge in the field to dynamically building a knowledge system in the field of building energy efficiency is the key issue of this paper.Facing this problem,few scholars currently quantify the knowledge development in the field of inquiry from the new perspective of bibliographic data mining.Based on this,this paper adopts secientometric method to construct a knowledge system in the field of building energy efficiency.According to the theoretical framework of disciplinary knowledge network,quantitatively mining knowledge network by scientometric method,this paper finally divides the knowledge from knowledge structure and knowledge path to identify knowledge subjects and knowledge trends to extract knowledge keywords.The research framework of a new type of knowledge system based on the new perspective of literature data mining in the field of building energy efficiency is introduced.Furthermore,on the basis of determining the literature search database and search terms,through cleaning and selecting the literature data,29,580 articles in the field of building energy efficiency have been collected on the Web of Science database.Based on these literature text data,this thesis conducts in-depth research from three aspects of the new knowledge system:First of all,in the study of the dividing knowledge structure of the field of building energy efficiency,we analyze the co-citation network in scientometric.By using the VOSviewer software’s own algorithm to cluster the co-citation network,based on the network characteristics of the subnet and the cited frequency of the node documents,the key documents that can reflect the main knowledge of each cluster are extracted.The qualitative interpretation of key documents shows that the five knowledge areas under the knowledge structure are: building energy efficiency knowledge area,phase change material knowledge area,life cycle knowledge area,building envelope knowledge area,and urban heat island knowledge area.Secondly,in the study of knowledge path identification knowledge subject of building energy efficiency,we research by the method of main path analysis in scientometric.As the foundation of main path analysis,scholars have rarely been studied the construction of a direct citation network,the construction method of this network is given here: after using Histcite and CRExplorer software to standardize the document data,the DOI number is used as a matching condition to use Matlab to prepare an algorithm program for calculating the citation matrix.Based on the visualized direct citation network of Pajek software,the main path analysis is used to calculate and extract branch knowledge paths in the five major knowledge areas,and the qualitative interpretation of the path literature can identify different knowledge in each knowledge area theme.Finally,in the research of extracting knowledge keywords of building energy efficiency from the knowledge trend,the co-word analysis and word frequency analysis in scientometric are used as the research methods.Due to the unique knowledge of disciplines in the field of building energy efficiency,these two methods are not applicable in the trend prediction research of this field.Therefore,the association strength index for extracting the key sets of knowledge topics is proposed to realize the effective overlap application of these two methods and has formed a unique knowledge trend prediction model.Using this model,it can be known that the thermal properties of phase change materials,energy consumption prediction,and life cycle assessment are the fastest-growing knowledge subjects.Based on the research on the above content,this paper analyzes the knowledge connotation from three dimensions of macro-meso-micro,and forms a new type of knowledge system based on quantitative data mining.The conclusions can provide valuable information on the overall knowledge development in the field for scholars.In addition,from the perspective of knowledge services for scholars,this paper gives a concrete idea of building a subject knowledge database in the field of building energy conservation in the future in the prospect section.