| 267 | 1 | 455 |
| 下载次数 | 被引频次 | 阅读次数 |
【目的】强对流天气通常具有突发性和局地性。有效获取社交媒体中的灾情信息，能够弥补强对流天气实况观测密度不足的问题，为灾害应急管理提供信息支撑。【方法】对典型强对流天气数据文本进行中文分词和统计分析，融合气象专家知识，形成适用于强对流天气的灾害主题语料库。将语义信息融入主题模型(LDA)和支持向量机(SVM)分类算法，构建了强对流灾害天气灾情信息提取模型。以江苏2021年“4·30”强风雹天气为例，收集了16 334条原创微博文本信息进行仿真试验。【结果】结果显示：(1)构建的强对流灾情信息提取模型对微博文本中的灾情信息识别与分类效果显著。一次文本主题挖掘提取出天气状况、科普防御、灾情影响、求助救援和其他信息5个主题，对“灾情影响”信息进行二次分类，提取出公共设施、电力通信、车辆交通、农业设施、人员伤亡和其他6种具体灾情信息。经过交叉验证，一次分类平均准确率为92.70%,二次分类平均准确率为90.95%。(2)将强对流天气发生发展过程划分为预警期、突发期和灾后期3个阶段，各类信息均在突发期处于高值区。灾害突发期，公共设施、电力通信信息最多，灾后期，人员伤亡信息讨论度最高。(3)灾情信息数量的空间分布基本与灾害影响严重地区一致。强对流天气发生时，公共设施暴露风险较高，受损现象最普遍。【结论】基于微博的强对流灾情信息提取模型能够有效获取微博文本隐含的灾情信息，反映灾害事件的变化特征和舆论焦点，对灾害监测预警服务和应急决策指挥具有一定参考意义。
Abstract：[Objective]Severe convective weather is typically characterized by abrupt onset and localized impact. Effectively acquiring disaster information from social media can compensate for the insufficient observation density of severe convective weather and provide information support for disaster emergency management.[Methods]Chinese text segmentation and statistical analysis were performed on text data of typical severe convective weather. By integrating meteorological expert knowledge, a disaster-themed corpus tailored to severe convective weather was developed. Semantic information was incorporated into the latent Dirichlet allocation(LDA) topic model and the support vector machine(SVM) classification algorithm to construct a disaster information extraction model under severe convective weather. Taking the severe wind-hail event in Jiangsu on April 30, 2021 as an example, 16 334 original Weibo text messages were collected for simulation experiments.[Results](1) The constructed disaster information extraction model for severe convective weather demonstrated remarkable effectiveness in identifying and classifying disaster information in Weibo texts. Through primary topic mining, five themes were extracted: weather conditions, public education on disaster prevention, disaster impact, rescue requests, and other information. The secondary classification was performed on “disaster impact” to extract six specific categories: public facilities, power and communication, vehicle traffic, agricultural facilities, casualties, and others. Cross-validation revealed an average accuracy of 92.70% for the primary classification and 90.95% for the secondary classification.(2) The development process of severe convective weather was divided into three stages: warning stage, outbreak stage, and post-disaster stage. All information categories peaked during the outbreak stage. During the outbreak stage, information on public facilities and power and communication was the most prevalent, while during the post-disaster stage, discussions about casualties were the most frequent.(3) The spatial distribution of disaster information quantity was generally consistent with the regions severely affected by the disaster. During severe convective weather events, public facilities faced a higher risk of exposure with damage being the most common.[Conclusion]The extraction model of disaster information based on Weibo for severe convective weather can effectively extract implicit disaster information in Weibo texts, reflect the variation characteristics of disaster events and the focus of public opinion, and provide valuable reference for disaster monitoring and early warning services as well as emergency response command.
[1] TANG J T,YANG S N,WANG W P.Social media-based disaster research:Development,trends,and obstacles[J].International Journal of Disaster Risk Reduction,2021,55:102095.
[2] YANG T F,XIE J B,LI G Q,et al.Social media big data mining and spatio-temporal analysis on public emotions for disaster mitigation[J].International Journal of Geo-Information,2019,8(1):29.
[3] WU K J,WU J D,LI Y.Mining typhoon victim information based on multi-source data fusion using social media data in China:A case study of the 2019 Super Typhoon Lekima[J].Geomatics,Natural Hazards and Risk,2022,13(1):1087-1105.
[4] AHMED Y A,AHMAD M N,AHMAD N,et al.Social media for knowledge-sharing:A systematic literature review[J].Telematics and Informatics,2019,37:72-112.
[5] WANG Z Y,YE X Y,TSOU M H.Spatial,temporal,and content analysis of Twitter for wildfire hazards[J].Natural Hazards,2016,83(1):523-540.
[6] 周义棋，田向亮，钟茂华.基于微博数据的自然灾害应急救助需求评估[J].清华大学学报(自然科学版),2022,62(10):1626-1635.ZHOU Yiqi,TIAN Xiangliang,ZHONG Maohua.Assessment of natural disaster emergency relief based on Microblog data,China[J].Journal of Tsinghua University(Science and Technology),2022,62(10):1626-1635.
[7] 张谱，张豪，孔锋，等.基于微博数据的暴雨洪涝灾害舆情特征研究：以 2021年中国三场暴雨洪涝为例[J].水利水电技术(中英文),2023,54(2):47-59.ZHANG Pu,ZHANG Hao,KONG Feng,et al.A study on public opinion characteristics of rainstorm flooding disasters based on Sina Weibo data:Take the three rainstorm flooding disasters in China in 2021 as an example[J].Water Resources and Hydropower Engineering,2023,54(2):47-59.
[8] United Nationgs.Sendai framework for disaster risk reduction 2015—2030[EB/OL].(2015-03-18) [2024-05-02].https://www.undrr.org/media/16176/download?startDownload=20240502.
[9] 陈梓，高涛，罗年学，等.反映自然灾害时空分布的社交媒体有效性探讨[J].测绘科学，2017,42(8):44-48.CHEN Zi,GAO Tao,LUO Nianxue,et al.Empirical discussion on relation between realistic disasters and social media data[J].Science of Surveying and Mapping,2017,42(8):44-48.
[10] CROOKS A,CROITORU A,STEFANIDIS A,et al.Earthquake:Twitter as a distributed sensor system[J].Transactions in GIS,2013,17(1):124-147.
[11] SMITH L,LIANG Q,JAMES P,et al.Assessing the utility of social media as a data source for flood risk management using a real-time modelling framework[J].Journal of Flood Risk Management,2017,10(3):370-380.
[12] 高志国，李毅，张利辉，等.洪水灾害时空过程模拟可视化表达研究进展与展望[J].水利水电技术(中英文),2023,54(8):43-53.GAO Zhiguo,LI Yi,ZHANG Lihui,et al.Research overview on spatiotemporal process simulation and expression of flood disaster scenarios[J].Water Resources and Hydropower Engineering,2023,54(8):43-53.
[13] TANG Z H,ZHANG L G,XU F H,et al.Examining the role of social media in California’s drought risk management in 2014[J].Natural Hazards,2015,79(1):171-193.
[14] SACHDEVA S,MCCAFFREY S,LOCKED.Social media approaches to modeling wildfire smoke dispersion:Spatio-temporal and social scientific investigations[J].Information,Communication & Society,2016,20 (8):1146-1161.
[15] WANG Y,RUAN S,WANG T,et al.Rapid estimation of an earthquake impact area using a spatial logistic growth model based on social media data[J].International Journal of Digital Earth,2019,12 (11):1265-1284.
[16] 薄涛.基于社交媒体的地震灾情数据挖掘与烈度快速评估应用[D].哈尔滨：中国地震局工程力学研究所，2018.BO Tao.Earthquake Disaster Data Mining and Application of Rapid Intensity Assessment based on Social Media[D].Harbin:Institute of Engineering Mechanics,China Earthquake Administration,2018.
[17] KRYVASHEYEU Y,CHEN H,OBRADOVICH N,et al.Rapid assessment of disaster damage using social media activity[J].Science Advances,2016,2(3):e1500779.
[18] 苏凯，程昌秀，Murzintcev N,等.主题模型在基于社交媒体的灾害分类中的应用及比较[J].地球信息科学学报，2019,21(8):1152-1160.SU Kai,CHENG Changxiu,Murzintcev N,et al.Application and comparison of topic model in identifying latent topics from disaster-related tweets[J].Journal of Geo-Information Science,2019,21(8):1152-1160.
[19] HUANG S,DU Y Y,YI J W,et al.Understanding human activities in response to typhoon hato from multi-source geospatial big data:A case study in Guangdong,China[J].Remote Sensing,2022,14(5):1269.
[20] XIAO C,ZHANG X D,XING Z Y,et al.Investigation of the Expression Method of Theme-Typhoon Disaster Information [J].ISPRS International Journal of Geo-Information,2021,10(3):109.
[21] CHEN Z,LIM S.Social media data-based typhoon disaster assessment [J].International Journal of Disaster Risk Reduction,2021,64:102482.
[22] 谢雪苗，邵亦文.社交媒体数据分析在台风灾害管理中的应用潜力探究：以台风“杜苏芮”对福建省的影响为例[J].热带地理，2024,44(6):1090-1101.XIE Xuemiao,SHAO Yiwen.Application potential of social media data analytics in typhoon disaster management:Taking the impact of Typhoon Doksuri on Fujian province as an example[J].Tropical Geography,2024,44(6):1090-1101.
[23] 梁春阳，林广发，张明锋，等.社交媒体数据对反映台风灾害时空分布的有效性研究[J].地球信息科学学报，2018,20(6):807-816.LIANG Chunyang,LIN Guangfa,ZHANG Mingfeng,et al.Assessing the effectiveness of social media data in mapping the distribution of typhoon disasters[J].Journal of Geo-Information Science,2018,20(6):807-816.
[24] 杨辰，潘顺，严岩.基于自然语言识别的上海市报警灾情数据识别及其气象灾害特征分析研究[J].自然灾害学报，2021,30(3):142-150.YANG Chen,PAN Shun,YAN Yan.Research on recognition of alarm disaster data and analysis of meteorological disaster features in Shanghai based on natural language recognition algorithm[J].Journal of Natural Disasters,2021,30(3):142-150.
[25] 王艳东，李昊，王腾，等.基于社交媒体的突发事件应急信息挖掘与分析[J].武汉大学学报(信息科学版),2016,41(3):290-297.WANG Yandong,LI Hao,WANG Teng,et al.The mining and analysis of emergency information in sudden events based on social media[J].Geomatics and Information Science of Wuhan University,2016,41(3):290-297.
[26] 黄晶，李梦晗，康晋乐，等.基于社交媒体的暴雨灾情信息实时挖掘与分析：以2019年“4·11深圳暴雨”为例[J].水利经济，2021,39(2):86-94.HUANG Jing,LI Menghan,KANG Jinle,et al.Mining and analysis of rainstorm disaster information based on social media—Case study of Shenzhen rainstorm on April 11,2019[J].Journal of Economics of Water Resources,2021,39(2):86-94.
[27] CHAUHAN U,SHAH A.Topic modeling using Latent Dirichlet Allocation [J].ACM Computing Surveys,2021,54(7):1-35.
[28] 顾荣直，吴洪颜，吴海英.基于灰色关联度等方法的江苏雷暴大风灾害风险评估与区划[J].暴雨灾害，2025,44(1):91-99.GU Rongzhi,WU Hongyan,WU Haiying.Risk assessment and zoning of thunderstorm-gale disasters in Jiangsu Province based on grey correlation degree and other methods [J].Torrential Rain and Disasters,2025,44(1):91-99.
[29] 陈静雯，马福民，刘新，等.基于神经网络的预警领域分词仿真算法[J].计算机仿真，2021,38(12):1-6.CHEN Jingwen,MA Fumin,LIU Xin,et al.Word segmentation simulation algorithm based on neural network in early warning field [J].Computer Simulation,2021,38(12):1-6.
[30] BLEI D M,NG A Y,JORDAN M I,et al.Latent Dirichlet Allocation[J].Journal of Machine Learning Research,2003,3:993-1022.
[31] 刘丽华.自然灾害微博舆情的社会计算模型建构研究[D].武汉：武汉大学，2018.LIU Lihua.Research on Social Computing Model Construction of Natural Disasters Weibo Public Opinion based on Sentiment Analysis and Topic Modeling[D].Wuhan:Wuhan University,2018.
[32] HANKAR M,KASRI M,HSSANE B A.A comprehensive overview of topic modeling:Techniques,applications and challenges[J].Neurocomputing,2025,628:129638.
[33] JELODAR H,WANG Y L,YUAN C,et al.Latent Dirichlet allocation (LDA) and topic modeling:Models,applications,a survey[J].Multimedia Tools and Applications,2019,78(11):15169-15211.
[34] CORTES C,VAPNIKV.Support Vector Machine[J].Machine Learning,1995,20(3):273-297.
[35] OYEDELE O.Determining the optimal number of folds to use in a K-fold cross-validation:A neural network classification experiment[J].Research in Mathematics,2023,10(1):2201015.
[36] NTI I K,NYARKO-BOATENG O,ANING J.Performance of machine learning algorithms with different K values in K-fold cross-validation[J].International Journal of Information Technology and Computer Science,2021,13(6):11.
[37] ZHANG X Y,LIU C A.Model Averaging prediction by K-Fold Cross-Validation[J].Journal of Econometrics,2023,235(1):280-301.
[38] 陈圣劼，刘梅，杨梦兮，等.江苏“4·30”强风雹成因及双偏振雷达特征分析[J].气象科学，2022,42(5):638-649.CHEN Shengjie,LIU Mei,YANG Mengxi,et al.Analysis on causes of ‘4.30’ severe gales and hails event and associated characteristics of dual-polarization radar echoes over Jiangsu[J].Journal of the Meteorological Sciences,2022,42(5):638-649.
[39] 杨永清，王鹏博，张媛媛.气象灾害舆情的时空演化特征及影响因素分析[J].晋图学刊，2024(4):16-29.YANG Yongqing,WANG Pengbo,ZHANG Yuanyuan.Analysis of spatio-temporal evolution characteristics and influencing factors of public opinion during meteorological disasters[J].Shanxi Library Journal,2024(4):16-29.
[40] 刘淑涵，王艳东，付小康.利用卷积神经网络提取微博中的暴雨灾害信息[J].地球信息科学学报，2019,21(7):1009-1017.LIU Shuhan,WANG Yandong,FU Xiaokang.Extracting rainstorm disaster information from microblogs using convolutional neural network[J].Journal of Geo-information Science,2019,21(7):1009-1017.
[41] 黄晶，吴星妍，李梦晗.不同主体视角下极端暴雨灾害事件网络舆情演化研究[J].水利经济，2023,41(4):94-101.HUANG Jing,WU Xingyan,LI Menghan.Research on the evolution of network public opinion of extreme rainstorm disaster events from the perspective of different subjects[J].Journal of Economics of Water Resources.2023,41(4):94-101.
[42] 张晓蔚，李佳帅，朱亮，等.南通机场4.30大风天气过程诊断分析[J].民航学报，2023,7(3):59-64.ZHANG Xiaowei,LI Jiashuai,ZHU Liang,et al.Diagnosis and analysis of a gale weather process at Nantong airport[J].Journal of Civil Aviation,2023,7(3):59-64.
[43] 吴洪颜，顾荣直，王勇.基于灾情数据的江苏省大风灾害等级和损失风险评价方法[J].气象科学，2024,44(5):997-1002.WU Hongyan,GU Rongzhi,WANG Yong.Evaluation method of gale disaster grade and loss risk in Jiangsu Province based on disaster data[J].Journal of the Meteorological Sciences,2024,44(5):997-1002.
[44] LEYKIN D,LAHAD M,AHARONSON-DANIELL.Gauging urban resilience from social media[J].International Journal of Disaster Risk Reduction,2018,31:393-402.
[45] ANJUM S,MUSTAQ Z,SARWAR M,et al.Climate change and its impact on agriculture[J].Journal of Agriculture and Biology,2024,2(2):199-218.
[46] FAN C,ESPARZA M,DARGIN J,et al.Spatial biases in crowdsourced data:Social media content attention concentrates on populous areas in disasters [J].Computers,Environment and Urban Systems,2020,83:101514.
[47] JAIDKA K,GUNTUKU S C,LEE J H,et al.The rural-urban stress divide:Obtaining geographical insights through Twitter [J].Computers in Human Behavior,2021,114:106544.
[48] PALA,WANG J,WU Y,et al.Social media driven big data analysis for disaster situation awareness:A tutorial [J].IEEE Transactions on Big Data,2023,9(1):1-21.
[49] 张岚，艾文文，罗晓春，等.气象灾害下网络信息传播模式研究：以台风为例[J].气象科学，2020,40(6):868-874.ZHANG Lan,AI Wenwen,LUO Xiaochun,et al.Research on communication of network information under meteorological disaster:Case of typhoon[J].Journal of the Meteorological Sciences,2020,40(6):868-874.
基本信息:
DOI：10.13928/j.cnki.wrahe.2025.10.001
中图分类号:TP391.1;P429
引用信息:
[1]张岚,李娟,陈静雯,等.基于微博的强对流天气灾情信息提取与分析：以江苏2021年“4·30”强风雹天气为例[J].水利水电技术(中英文),2025,56(10):1-16.DOI:10.13928/j.cnki.wrahe.2025.10.001.
基金信息:
国家自然科学基金项目(42171081); 中国气象局创新发展专项(CXFZ2024J012); 江苏省气象局面上项目(KM202205)
2025-08-22
2025-08-22
2025-08-22