中国标准化Issue(5)：61-65,5.DOI:10.3969/j.issn.1002-5944.2024.05.008
基于深度学习的法人和其他组织国民经济行业分类标准化流程研究
Research on Standardized Process of Industrial Classification for National Economic Activities for Legal Entities and Other Organizations Based on Deep Learning
摘要
Abstract
Aiming at the problems of low efficiency and accuracy in the industrial classification of national economic activities of legal entities and other organizations,a standardized process of automatic Classification of industrial data of national economic activities based on BERT-LSTM-CNN model is proposed.First,the accuracy of the input data is ensured by quality assessment and adjustment of the Unified code data.Second,a well-trained BERT-LSTM-CNN hybrid model is used to realize feature extraction,and self-attention mechanism and transfer learning strategy are applied to effectively deal with the industrial Classification problem.This standardized process not only provides users in various industries with accurate,timely and comprehensive industrial Classification information of national economic activities,but also provides solid data support for the decision-making process.关键词
法人和其他组织/国民经济行业分类/标准化流程/深度学习Key words
legal entity and other organization/industrial classification for national economic activities/standardized process/deep learning引用本文复制引用
袁辉,赵捷,侯博,李晟飞,韩雪..基于深度学习的法人和其他组织国民经济行业分类标准化流程研究[J].中国标准化,2024,(5):61-65,5.基金项目
本文受国家市场监督管理总局科技计划项目"基于深度学习技术的法人和其他组织国民经济行业分类机器判定研究"(项目编号:2020MK185)资助. （项目编号:2020MK185）
