Research And Implementation Of The Establishment Method Of Science And Technology Entry Database

With the rapid development of science and technology,many new technical fields and research results are increasing day by day,which brings a lot of difficulties to the analysis of technical fields.The current definition of scientific and technological research content is mainly through classification methods such as disciplines and technical fields.This coarse-grained classification method limits the development of scientific research big data analysis and the exchange of scientific and technological information.In response to this problem,the research group put forward the concept of technology entry to describe science and technology theories and technical research categories,and developed a technology entry platform and system that integrates resources related to science and technology research and services based on technology entry.Technology entries are derived from scientific research literature.This thesis studies the storage of science and technology entries.The main work of this thesis is as follows:(1)A new word discovery method based on information entropy and point mutual information technology is proposed.According to the statistical characteristics of technology entries in scientific and technological literature,this thesis uses a combination of information entropy and point mutual information to extract technical entry new words.When extracting technology candidate words,information entropy and point mutual information are used to calculate the internal aggregation strength of the string,so as to ensure that the string is a legal language unit and establish a candidate word set for technology entries.The experimental result has an accuracy rate of 35.9%,indicating that the algorithm can effectively extract scientific and technological terms,and after using a negative sample set,the accuracy rate is increased to 55.7%.(2)On the basis of information entropy and point mutual information,combined with the language model in scientific research literature,a new technology word discovery algorithm based on language mode,information entropy and point mutual information is proposed.First,the basic entry set is extracted by the method based on information entropy and point mutual information.Secondly,the language model is used to extract the candidate technology entry set and compare it with the basic entry set.Then delete the wrong entries in the candidate technology entry set,and finally get the new technology entry set.The experimental result has an accuracy rate of 79%,indicating that the algorithm can effectively extract scientific and technological entries from scientific and technological literature.(3)Designed and implemented a technology entry database system.First,analyze the requirements of the scientific and technological term database system,divide the functional modules of the scientific and technological term database system,and secondly design the core function modules of the system in detail,and design the key database tables.Finally,this thesis implements the technology entry database system.It provides functions such as discovery of new science and technology words,storage of science and technology terms,management of technology entry,and navigation of science and technology resources,so as to achieve a high degree of integration of research objects,research results,research subjects and related resources of the achievement industrialization chain.