Traceability Of Graphene Preparation And The Research And Implementation Of Detection Standard Data Management System

With the development of graphene-related technologies,the relevant literature published by researchers has shown exponential growth.It is very difficult for researchers to find a suitable literature facing a large amount of scientific research literature.Secondly,it is user’s eager hope to quickly select high-quality literature from uneven literature.Meanwhile,the related data of graphene is distributed on different websites,and it is very difficult to search.It needs to be integrated to save the query time of relevant researchers.Finally,researchers need to understand the evolution process of graphene preparation methods,and then classify and trace them.In order to solve these problems,this project is in cooperation with the National Graphene Product Quality Supervision and Inspection Center.This Project has developed a graphene data information management system with literature search function for graphene and literature classification function for graphene.In this paper,the full-text search technology Lucene is used to realize the rapid retrieval of graphene literature and the related shallow machine learning and deep learning algorithms are used to classify graphene literature.The main research work is as follows:First,optimizing the algorithm of Lucene’s search result to make the search results related to the quality of the literature.After Researching the search ranking algorithm of existing search engines,this paper proposes a literature quality calculation algorithm.Then this paper combines it with Lucene default correlation algorithm to improve the accuracy of searching literature related to graphene and provide more objective and fair search results of literature.Then,in order to realize the classification function of graphene literature,this paper studies the existing techniques based on shallow machine learning and deep learning,and on this basis,this paper optimizes the feature selection algorithm.This paper applies the new algorithm on different classification algorithms and compare the final classification accuracy,and then choose the suitable classification algorithm.Finally,in order to provide above functions to researchers to use,this paper finally designed a graphene information management system for the literature search system,classification system and a big data storage system for storing a large amount of literature.The experimental results show that the literature quality algorithm is real and effective,and the 3-times optimization algorithm for Lucene improves the total quality of retrieval results.The optimized text classification algorithm improves the accuracy of text classification.The big data storage system effectively solves the problem of storing and accessing a large number of literature.The graphene information management system effectively improves the efficiency of researchers.