人大报告内容的文本分类  

Text classification of NPC report contents

作　　者：喻航 李红莲[1] 吕学强[2] YU Hang;LI Hong-lian;LYU Xue-qiang(School of Information and Communication Engineering,Beijing Information Science and Technology University,Beijing 100101,China;Beijing Key Laboratory of Internet Culture and Digital Dissemination Research,Beijing Information Science and Technology University,Beijing 100101,China)

机构地区：[1]北京信息科技大学信息与通信工程学院,北京100101 [2]北京信息科技大学网络文化与数字传播北京市重点实验室,北京100101

出　　处：《计算机工程与设计》2021年第6期1772-1778,共7页Computer Engineering and Design

基　　金：国家自然科学基金项目(61671070);北京成像技术高精尖创新中心基金项目(BAICIT-2016003);国家社会科学基金重大基金项目(14@ZH0363);国家语委重点基金项目(ZDI135-53)。

摘　　要：针对区级人大报告特定的几方面内容进行文本分类,可以让人大工作人员对不同工作内容进行快速分辨,是构建人大报告辅助生成系统的必要内容。为对不同内容分类,基于TF-IDF(词频-逆文档频率)与知识增强语义表示模型ERNIE(enhanced representation from knowledge integration)结合构建分类模型。ERNIE直接对语义知识单元进行建模,在此基础上加入TF-IDF提升模型性能。实验结果表明,该方法在分类的准确率和召回率上表现不错,使ERNIE模型收敛速度加快,通过该模型可以较好地对人大报告的文本进行分类。Text classification for specific aspects of district-level people’s congress reports can allow staff members of the congress to quickly distinguish between different work contents,and is a necessary content to build an auxiliary generation system for people’s congress reports.To classify different contents,a classification model based on the combination of TF-IDF(term frequency-inverse document frequency)and knowledge-enhanced semantic representation model ERNIE(enhanced representation from knowledge integration)was proposed.ERNIE directly modeled the semantic knowledge unit.On this basis,adding TF-IDF improved the performance of the model.Experimental results show that the proposed method performs well in classification accuracy and recall rate,which accelerates the convergence speed of the ERNIE model.The model can be used to better classify the text of the NPC report.

关 键 词：人大报告 文本分类 词频-逆文档频率 知识增强语义表示模型 速度 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]