Research on Standard Literature Classification Based on Large Language Model
In today's era of big data,the explosive growth of standards and other literature poses significant challenges for the efficient management and services of documents.Due to the continuous evolution and diversification of industries,traditional standard classification systems struggle to adapt flexibly to the ever-changing demands of industries,resulting in a gap between standard classification and actual industrial need.In the information age,this problem is notably emphasized,and the transformation and upgrading of traditional standard classifications become challenging.Therefore,addressing the issue of standard classifications being difficult to align with industrial needs has become a crucial aspect of enhancing the efficiency and quality of document management and services.In this context,the innovative approach proposed in this paper aims to bridge the gap between standard classification and industries,enhancing the accuracy of industrial classification to better meet the evolving demands of industries.Simultaneously,this method focuses on addressing the complex challenges faced in the field of Chinese industrial classification,including issues such as multiple semantics,multiple categories,and limited annotated data.
large language modelssemantic representationliteraturestandard
万方数据
维普
