民航旅客不文明行为信息自动匹配方法  

Automatic matching method of civil aviation passenger uncivilized behavior information

作　　者：曹卫东[1] 高德 CAO Wei-dong;GAO De(School of Computer Science and Technology,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学计算机科学与技术学院,天津300300

出　　处：《计算机工程与设计》2021年第12期3415-3421,共7页Computer Engineering and Design

基　　金：民航科技创新重大专项基金项目(MHRD20160109);民航安全能力基金项目(TRSA201803);国家自然基金民航联合基金项目(U1833114)。

摘　　要：为实现民航旅客不文明行为信息与规则自动匹配,提出一种将胶囊网络与门控循环单元结合的民航旅客不文明行为信息与规则自动匹配的方法。充分利用文本语义信息,进行文本相似度的计算,有效解决传统方法在民航旅客不文明行为信息中存在繁杂、无用信息的干扰导致效率低且需大量人工成本的问题。通过在两个公共数据集和民航旅客不文明信息数据集上的实验,验证了该设计的准确性和有效性,为民航旅客不文明行为信息与规则的自动匹配提供了方法支持。To realize the automatic matching of civil aviation passengers uncivilized behavior information and rules,a method for automatically matching civil aviation passengers uncivilized behavior information and rules was proposed,in which a capsule network and a gated recurrent unit were combined.The GRU-capsule combination model made full use of text semantic information to calculate text similarity,and effectively solved the interference of complex and useless information in the civil aviation passengers uncivilized behavior information,resulting in low efficiency and requiring a lot of labor costs.Through experiments on two public data sets and civil aviation passenger uncivilized information data sets,the accuracy and effectiveness of the design are verified.It provides method support for the automatic matching of civil aviation passengers uncivilized behavior information and rules.

关 键 词：不文明旅客 文本相似度 门控循环单元 胶囊网络 集成模型 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]