计算机与数字工程2025，Vol.53Issue(4)：1044-1050,7.DOI:10.3969/j.issn.1672-9722.2025.04.022
基于混合神经网络的政务热线事件提取方法
Event Extraction Method for Government Hotline Based on Hybrid Neural Network
摘要
Abstract
A work-order event extraction method based on hybrid neural network is proposed to address the problems of low ef-ficiency with manual classification owing to the increasing government hotline traffic.The method takes into account the local and global features of the work-order text via CNN and BiGRU network,processes the global features through the self-attention mecha-nism to highlight the semantic features with high importance,and assigns tags to each word and calculates the tag sequence score of the work-order text to enhance the event extraction effect of the work-order.At the same time,the method also introduces memory network to solve the problem of sparse content for the government hotline work-order text.Comparative experimental results show that the method can achieve better performance than the baseline methods.关键词
政务热线/事件提取/记忆网络/混合神经网络Key words
government hotline/event extraction/memory network/hybrid neural network分类
信息技术与安全科学引用本文复制引用
强婷婷,陈剑,佘祥荣,陈健鹏,陈钢..基于混合神经网络的政务热线事件提取方法[J].计算机与数字工程,2025,53(4):1044-1050,7.基金项目
国家自然科学基金项目(编号:61976198) （编号:61976198）
安徽省重点研究与开发计划(编号:202104a05020071)资助. （编号:202104a05020071）
