{"title":"A Non-standardized Chinese Express Delivery Address Identification Model Based on Enhanced Representation","authors":"Zi Ye, Xuefeng Piao, F. Meng, Bo Cao, Dianhui Chu","doi":"10.1109/ICSS55994.2022.00011","DOIUrl":null,"url":null,"abstract":"Intelligent recognition of express delivery address information is an important means to improve the efficiency of express list filling. At present, the confusion in user input and the inconsistent expression brings challenges to intelligent address information recognition. In addition, the recognition accuracy of the existing solutions has low identification and limited by the data format. To address the problems, this paper proposes a non-standardized Chinese express delivery address identification model based on enhanced representation, which has been improved from two aspects: entity extraction and memory network. The improved entity extraction model based on word embedding can mine the context information in the text in both positive and negative directions and consider the correlation between characters, thus outputting a more accurate prediction sequence. In order to solve the problem of information overload in existing models, a convolutional block memory network based on BERT was designed. The results of the experimental comparative analysis showed that the improved method effectively improved the identification accuracy of non-standardized Chinese express delivery address information, thus proving the effectiveness and availability of the method.","PeriodicalId":327964,"journal":{"name":"2022 International Conference on Service Science (ICSS)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 International Conference on Service Science (ICSS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICSS55994.2022.00011","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Intelligent recognition of express delivery address information is an important means to improve the efficiency of express list filling. At present, the confusion in user input and the inconsistent expression brings challenges to intelligent address information recognition. In addition, the recognition accuracy of the existing solutions has low identification and limited by the data format. To address the problems, this paper proposes a non-standardized Chinese express delivery address identification model based on enhanced representation, which has been improved from two aspects: entity extraction and memory network. The improved entity extraction model based on word embedding can mine the context information in the text in both positive and negative directions and consider the correlation between characters, thus outputting a more accurate prediction sequence. In order to solve the problem of information overload in existing models, a convolutional block memory network based on BERT was designed. The results of the experimental comparative analysis showed that the improved method effectively improved the identification accuracy of non-standardized Chinese express delivery address information, thus proving the effectiveness and availability of the method.