摘要：

Abstract:

【Purpose/significance】By analyzing the institutions, theory and judicial practice of copyright infringement prevention in

digital libraries, the core controversial issues and improvement paths to prevent copyright infringement are clarified

.

【Method/process】

Taking chinalawinfo.com and CNKI as data sources, we selected laws and regulations, policy documents, authoritative academic pa⁃

pers and typical case judgments as research samples, and comprehensively applied the methods of Keyword Co-Occurrence, Domain

Academic Claim Base System, DTM analysis, Text Comparison analysis and Word2vec analysis to carry out the correlation analysis.

【Result/conclusion】Through the correlation analysis and test, it is found that the subjects concerned have different degrees of cogni⁃

tive bias towards the reasonable use of digital works, artificial intelligence brings an important opportunity to build a copyright system

with Chinese characteristics, and the function of the copyright collective management system has become the key to the prevention of

copyright infringement.【Innovation/limitation】With the help of the method of constructing a library of academic views in the field to

assist text analysis, it has an upgrading effect on the separate use of text analysis methods, and has an innovative value in the study of

perfecting the path of preventing copyright infringement in digital libraries. Since correlation research is a complex task, the results

will be affected by the quality of data, pre-processing methods and the chosen methods. The construction of the Domain Academic

Claim Base System is subject to further rationality and feasibility verification, which may bring about minor uncertainties.