面向飞机故障文本的信息抽取  

Information Extraction for Aircraft Fault Text

作　　者：乔璐 孙有朝[1] 吴红兰[1] QIAO Lu;SUN You-chao;WU Hong-lan(College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 210000,China)

机构地区：[1]南京航空航天大学民航学院,江苏南京210000

出　　处：《计算机与现代化》2024年第3期61-66,71,共7页Computer and Modernization

基　　金：国家自然科学基金委员会-中国民用航空局民航联合研究基金资助项目(U2033202,U1333119);国家自然科学基金资助项目(52172387)

摘　　要：针对人工提取飞机故障信息工作量大、效率低、成本高等问题,提出一种基于领域词典、规则和BiGRU-CRF模型的信息抽取方法。结合飞机领域知识的特点,基于飞机故障文本信息构建领域词典库和模板规则,并对故障信息进行语义标注。采用BiGRU-CRF深度学习模型进行命名实体识别,BiGRU获取上下文的语义关系,CRF解码生成实体标签序列。实验结果表明,基于领域词典、规则和BiGRU-CRF模型的信息抽取方法准确率高达95.2%,验证了该方法的有效性。本文方法能够准确识别出飞机故障文本中的关键词如时间、机型、故障件名称、故障件制造单位等信息,同时,根据领域词典和规则对识别结果进行修正,有效提高了信息抽取的效率和准确性,解决了传统实体抽取模型长期依赖人工特征的问题。In view of the problems of large workload,low efficiency and high cost of manual extraction of aircraft fault informa⁃tion,a method of information extraction based on domain dictionary,rules and BiGRU-CRF model is proposed.Combining the characteristics of aircraft domain knowledge,domain dictionary and template rules are constructed based on aircraft fault text in⁃formation,and semantic labeling of fault information is carried out.The BiGRU-CRF deep learning model is used for named en⁃tity recognition.BiGRU obtaines the semantic relationship of context,and CRF decodes and generates the entity label sequence.The experimental results show that the information extraction method based on domain dictionary,rules and BiGRU-CRF model has an accuracy of 95.2%,which verifies the effectiveness of the method.It can accurately identify the key words in the aircraft fault text,such as time,aircraft type,fault part name,fault part manufacturer and other information.At the same time,accord⁃ing to the domain dictionary and rules to correct the recognition results,effectively improves the efficiency and accuracy of infor⁃mation extraction,and solves the problem of traditional entity extraction model long-term dependence on manual features.

关 键 词：故障信息 信息抽取 命名实体识别 BiGRU-CRF 领域词典 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]