计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (12): 30-36.doi: 10.11896/jsjkx.240300025

• 数字孪生网络与人工智能融合 • 上一篇    下一篇

基于深度对比孪生网络的事件辨重方法

李子琛1, 易修文2,3, 陈顺1,2,3, 张钧波1,2,3, 李天瑞1   

  1. 1 西南交通大学计算机与人工智能学院 成都 611756
    2 北京京东智能城市大数据研究院 北京 100176
    3 京东城市(北京)数字科技有限公司 北京 100176
  • 收稿日期:2024-03-04 修回日期:2024-07-17 出版日期:2024-12-15 发布日期:2024-12-10
  • 通讯作者: 易修文(xiuwenyi@foxmail.com)
  • 作者简介:(zichen_li@126.com)
  • 基金资助:
    国家重点研发计划(2023YFC2308703);北京市科技新星(Z211100002121119)

Deep Contrastive Siamese Network Based Repeated Event Identification

LI Zichen1, YI Xiuwen2,3, CHEN Shun1,2,3, ZHANG Junbo1,2,3, LI Tianrui1   

  1. 1 School of Computing and Artificial Intelligence, Southwest Jiaotong University, Chengdu 611756, China
    2 JD Intelligent Cities Research, Beijing 100176, China
    3 JD Intelligent Cities Technology Co., Ltd., Beijing 100176, China
  • Received:2024-03-04 Revised:2024-07-17 Online:2024-12-15 Published:2024-12-10
  • About author:LI Zichen,born in 1997,postgraduate.His main research interests include urban computing and deep learning.
    YI Xiuwen,born in 1991,Ph.D,data scientist,researcher,is a senior member of CCF(No.45025M).His main research interests include spatio-temporal data mining and deep learning.
  • Supported by:
    National Key R&D Program of China(2023YFC2308703) and Beijing Nova Program(Z211100002121119).

PDF (PC)

251

摘要： 在中国,市民可以通过拨打12345市民热线,向政府报告生活中遇到的问题并寻求帮助。然而,有许多重复的事件被多次上报,这给负责事件分派的工作人员带来了很大的压力,也会导致事件的处置效率变低,浪费社会公共资源。对重复事件的判断需要精确分析文本语义和上下文关系,为了解决这个问题,文中提出了一种基于深度对比孪生网络的事件辨重方法,通过评估两个事件的描述文本之间的相似性,辨别出具有相同诉求的事件。首先通过召回和过滤的方法来减少候选事件的数量;然后通过对比学习构造任务,微调预训练的BERT模型,学习易于辨识的事件描述语义表征;最后引入事件标题作为上下文信息,并通过带有分类器的孪生网络来识别重复事件。在南通市12345事件数据集上进行了实验,结果表明,该方法在各项评估指标上均优于基线方法,特别是在与辨重任务场景相关的F0.5分数上,能够有效地辨别重复事件,提高事件处置的效率。

关键词: 12345热线, 重复事件识别, 对比学习, 孪生网络, 城市计算

Abstract: In China,citizens can report issues they encounter in daily life to the government and seek assistance by calling the 12345 citizen hotline.However,many events are reported multiple times,which places significant pressure on the staffs responsible for event allocation,resulting in low efficiency of event disposal and waste of public resources.Identifying repeated events requires precise analysis of textual semantics and contextual relationships.To address this problem,this paper proposes an event repetition identification method based on a deep contrastive siamese network.By evaluating the similarity between the descriptions of events,the method identifies events with the same demands.First,it reduces the number of events through retrieval and filtering.Then,it fine-tunes a pre-trained BERT model through contrastive learning to learn distinct semantic representations of event descriptions.Finally,the event title is introduced as contextual information,and a siamese network with a classifier is used to identify repeated events.Experimental results on the 12345 event dataset of Nantong demonstrate that the proposed method outperforms baseline methods across various evaluation metrics,particularly in the F0.5 score,which is relevant to the repetition task scenario.The proposed method can effectively identify repeated events and improve the efficiency of event handling.

Key words: 12345 hotline, Repeated event dispatch, Contrastive learning, Siamese network, Urban computing

中图分类号: 

  • TP399
[1]DEVLIN J,CHANG M W,LEE K,et al.Bert:Pre-training ofdeep bidirectional transformers for language understanding[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics.Human Language Technologies,2019:4171-4186.
[2]SARZYNSKA-WAWER J,WAWER A,PAWLAK A,et al.Detecting formal thought disorder by deep contextualized word representations[J].Psychiatry Research,2021,304:114135.
[3]YANG Z,DAI Z,YANG Y,et al.Xlnet:Generalized autoregressive pretraining for language understanding[C]//Proceedings of the 33rd International Conference on Neural Information Processing Systems.2019:5753-5763.
[4]ZHENG Y P,MA X L.Using Government Hotline Data to Promote Smart Governance－The Case of Guangzhou Government Hotline[J].E-Government,2018(12):18-26.
[5]MA X L,ZHENG Y P,ZHANG C W.The Big Data Empowering Effect of Government Hotlines on City Governance Innovation:Value,Status and Issues[J].Documentation,Informaiton &Knowledge,2021,38(2):4-12.
[6]CHENG X M,CHEN G,CHEN J P,et al.RAVA:An Reinforced-Association-Based Method for 12345 Hotline Events Allocation[J].Journal of Chinese Information Processing,2022,36(10):155-166,172.
[7]PU X,LONG K,CHEN K,et al.A semantic-based short-textfast clustering method on hotline records in Chengdu[C]//2019 IEEE Intl Conf on Dependable,Autonomic and Secure Computing,Intl Conf on Pervasive Intelligence and Computing,Intl Conf on Cloud and Big Data Computing,Intl Conf on Cyber Science and Technology Congress.2019:516-521.
[8]PENG X,LI Y,SI Y,et al.A social sensing approach for everyday urban problem-handling with the 12345-complaint hotline data[J].Computers,Environment and Urban Systems,2022,94:101790.
[9]LUO J Y,QIU Z,XIE G Q,et al.Research on civic hotline complaint text classification model based on word2vec[C]//2018 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery.IEEE,2018:180-1803.
[10]CHANDRASEKARAN D,MAGO V.Evolution of semanticsimilarity－a survey[J].ACM Computing Surveys(CSUR),2021,54(2):1-37.
[11]MIKOLOV T,CHEN K,CORRADO G,et al.Efficient estimation of word representations in vector space[C]//Proceedings of the 1th International Conference on Learning Representations.2013.
[12]PENNINGTON J,SOCHER R,MANNING C D.Glove:Global vectors for word representation[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Proces-sing.2014:1532-1543.
[13]KIM Y.Convolutional Neural Networks for Sentence Classification[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing.2014:1746-1751.
[14]JOHNSON R,ZHANG T.Deep pyramid convolutional neural networks for text categorization[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics.2017:562-570.
[15]LIU P,QIU X,HUANG X.Recurrent neural network for text classification with multi-task learning[C]//Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence.2016:2873-2879.
[16]VASWANI A,SHAZEER N,PARMAR N,et al.Attention isall you need[C]//Proceedings of the 31st International Confe-rence on Neural Information Processing Systems.2017:6000-6010.
[17]KATHERINE L,DAPHNE L,NYSTROM A,et al.Deduplica-ting Training Data Makes Language Models Better[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics.2022:8424-8445.
[18]BIKASH G,LUCAS A,PETR K.Deduplication of ScholarlyDocuments using Locality Sensitive Hashing and Word Embeddings[C]//Proceedings of the Twelfth Language Resources and Evaluation Conference.2020:901-910.
[19]JAISWAL A,BABU A R,ZADEH M Z,et al.A survey on con-trastive self-supervised learning[J].Technologies,2020,9(1):2.
[20]CHEN T,KORNBLITH S,NOROUZI M,et al.A simpleframework for contrastive learning of visual representations[C]//International Conference on Machine Learning.PMLR,2020:1597-1607.
[21]GAO T,YAO X,CHEN D.SimCSE:Simple Contrastive Lear-ning of Sentence Embeddings[C]//Proceedings of the 2021 Conference on Empirical Methods in Natural Language Proces-sing.2021:6894-6910.
[22]WIETING J,BANSAL M,GIMPEL K,et al.Towards Universal Paraphrastic Sentence Embeddings[C]//Proceedings of the 4th International Conference on Learning Representations.2016.
[23]CHEN X,HE K.Exploring Simple Siamese RepresentationLearning[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2021:15750-15758.
[24]SONG Y,SHI S,LI J,et al.Directional Skip-Gram:ExplicitlyDistinguishing Left and Right Context for Word Embeddings[C]//Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies.2018:175-180.
[25]IYYER M,MANJUNATHA V,BOYD-GRABER J,et al.Deep Unordered Composition Rivals Syntactic Methods for Text Classification[C]//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing.2015:1681-1691.
[26]WANG L,YANG N,HUANG X,et al.Text Embeddings byWeakly-supervised Contrastive Pre-training[J].arXiv:2212.03533,2022.
[27]LIU Y,OTT M,GOYAL N,et al.Roberta:A Robustly Optimized BERT Pretraining Approach[J].arXiv:1907.11692,2019.
[28]REIMERS N,GUREVYCH I.Sentence-BERT:Sentence Em-beddings using Siamese BERT-Networks[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing.2019:3982-3992.
[29]BLEI D M,NG A Y,JORDAN M I.Latent dirichlet allocation[J].Journal of Machine Learning Research,2003,3(Jan):993-1022.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发