[1]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(03):142-150.[doi:10.13577/j.jnd.2021.0316]
 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].,2021,30(03):142-150.[doi:10.13577/j.jnd.2021.0316]
点击复制

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
分享到:

《自然灾害学报》[ISSN:/CN:23-1324/X]

卷:
30
期数:
2021年03期
页码:
142-150
栏目:
出版日期:
2021-06-28

文章信息/Info

Title:
Research on recognition of alarm disaster data and analysis of meteorological disaster features in Shanghai based on natural language recognition algorithm
作者:
杨辰 潘顺 严岩
上海市气象灾害防御技术中心, 上海 200030
Author(s):
YANG Chen PAN Shun YAN Yan
Shanghai Center for Meteorological Disaster Prevention Technology, Shanghai 200030, China
关键词:
110灾情LDATF-IDF特征分析
Keywords:
110 disastersLDATF-IDFfeature analysis
分类号:
X4;X16
DOI:
10.13577/j.jnd.2021.0316
摘要:
上海作为特大型城市,其城市运行极易受到气象条件影响。受制于传统的灾害上报和普查渠道,灾情数据量较为有限且时效性不强。研究基于110气象灾情数据,采用自然语言处理算法,对报警灾情描述信息进行分析提取,识别灾种、承灾体和影响程度等信息,并进一步分析了气象灾害的时空特征以及致灾的气象条件。结果表明,该方法可以较好地提取气象灾害的类别、承灾体及影响信息。近10a灾情分布年际差异较大,月分布以8月和10月为最多,小时分布呈现双峰特征,且工作日更为明显。暴雨和大风的灾情空间分布均以中心城区为最高,其次为紧邻中心城区的外环以内区域以及各区的城市副中心。结合气象资料分析,短时强降雨和长时间的较强降雨是引发暴雨灾害的重要因素,瞬时大风往往会引起大风灾害,而台风和持续性强降水是造成灾情大量爆发的主要原因。
Abstract:
As a megacity, the urban operations of Shanghai are extremely susceptible to weather conditions. Subject to traditional disaster reporting and census channels, the amount of disaster data is relatively limited and time-consuming. The study is based on 110 meteorological disaster data and uses natural language processing algorithms to analyze and extract alarm disaster description information, identify disaster types, disaster-bearing bodies, and impact degrees, and further analyze the spatial and temporal characteristics of meteorological disasters and the weather conditions that cause disasterc. The results show that the method can extract the meteorological disaster categories, disaster-bearing bodies and impact information. The distribution of disasters in the past 10 years varies greatly from year to year. Monthly distribution has the largest number in August and October. The hourly distribution has a double-peak characteristic, and more obvious in the working day. The spatial distribution of torrential rain and strong wind disasters is the highest in the central city, followed by the area within the outer ring close to the central city and the urban sub-centers in each district. Analysis with meteorological data, short-term heavy rainfall and long-term rainfall are important factors that trigger heavy rain disasters. Instantaneous high winds often cause strong wind disasters, and also typhoon and persistent heavy precipitation are the main reasons for the large number of disasters.

参考文献/References:

[1] 杨辰, 王强, 顾宇丹.上海市城市暴雨内涝评估建模及模拟研究[J].气象, 2017, 43(7):879-886. YANG Chen, WANG Qiang, GU Yudan. Study of Shanghai urban flooding assessment modeling and simulation[J]. Meteorological Monthly, 2017, 43(7):879-886. (in Chinese)
[2] 杨辰, 王强, 金雪欢, 等.基于GIS的上海市社区暴雨积涝模拟研究[J].自然灾害学报, 2015, 24(2):195-202. YANG Chen, WANG Qiang, JIN Xuehuan, et al. GIS-based simulation study on community rainstorm waterlogging in Shanghai Municipality[J]. Journal of Natural Disasters, 2015, 24(2):195-202. (in Chinese)
[3] 殷杰, 尹占娥, 许世远.上海市灾害综合风险定量评估研究[J].地理科学, 2009, 29(3):450-454. YIN Jie, YIN Zhane, XU Shiyuan. Urban-disasters composite risk assessment of Shanghai[J]. Scientia Geographica Sinica, 2009, 29(3):450-454. (in Chinese)
[4] 贺芳芳, 赵兵科.近30年上海地区暴雨的气候变化特征[J].地球科学进展, 2009, 24(11):1260-1267. HE Fangfang, ZHAO Bingke. The characteristics of climate change of torrential rains in Shanghai region in recent 30 years[J]. Advances in Earth Science, 2009, 24(11):1260-1267. (in Chinese)
[5] 虎雅琼, 边宇轩, 黄梦宇, 等.基于灾情信息的1981-2017年北京地区降雹特征[J].应用气象学报, 2019, 30(6):710-721. HU Yaqiong, BIAN Yuxuan, HUANG Mengyu, et al. Characteristics of hailstone distribution based on disaster information in Beijing from 1981 to 2017[J]. Journal of Applied Meteorological Science, 2019, 30(6):710-721. (in Chinese)
[6] 刘佼, 金琼, 韩照全.江苏省雷灾特征分析及灾情灰色关联评估[J].气象与环境科学, 2019, 42(2):97-103. LIU Jiao, JIN Qiong, HAN Zhaoquan. Analysis of lightning characteristics in Jiangsu province and grey correlation assessment of disaster simulation[J]. Meteorological and Environmental Sciences, 2019, 42(2):97-103. (in Chinese)
[7] 俞海洋, 李婷, 陈笑娟, 等.河北省近30年大风时空分布及成灾特征分析[J].灾害学, 2017, 32(2):59-63. YU Haiyang, LI Ting, CHEN Xiaojuan, et al. The spatio-temporal distribution and disaster characteristics analysis of the gale in Hebei province for nearly 30 years[J]. Journal of Catastrophology, 2017, 32(2):59-63. (in Chinese)
[8] 於琍, 徐影, 张永香.近25a中国暴雨及其引发的暴雨洪涝灾害影响的时空变化特征[J].暴雨灾害, 2018, 37(1):67-72. YU Li, XU Ying, ZHANG Yongxiang. Temporal and spatial variation of rainstorms and the impact of flood disasters due to rainstorm in China in the past 25 years[J]. Torrential Rain and Disasters, 2018, 37(1):67-72. (in Chinese)
[9] 殷启元, 范祥鹏, 陈绿文, 等.广东省雷电伤亡事故特征分析[J].自然灾害学报, 2019, 28(6):204-213. YIN Qiyuan, FAN Xiangpeng, CHEN Lvwen, et al. Characteristic analysis of Guangdong lightning casualties[J]. Journal of Natural Disasters, 2019, 28(6):204-213. (in Chinese)
[10] 薄涛, 李小军, 陈苏, 王玉婷, 等.基于社交媒体数据的地震烈度快速评估方法[J].地震工程与工程振动, 2018, 38(5):206-215. BO Tao, LI Xiaojun, CHEN Su, et al. Research of seismic intensity rapid assessment based on social media data[J]. Earthquake Engineering and Engineering Dynamics, 2018, 38(5):206-215. (in Chinese)
[11] 徐敬海, 褚俊秀, 聂高众, 等.基于位置微博的地震灾情提取[J].自然灾害学报, 2015, 24(5):12-18. XU Jinghai, CHU Junxiu, NIE Gaozhong, et al. Earthquake disaster information extraction based on location microblog[J]. Journal of Natural Disasters, 2015, 24(5):12-18. (in Chinses)
[12] 曹彦波.基于新浪微博的2018年云南通海5.0级地震舆情时空特征分析[J].地震研究, 2018, 41(4):525-533. CAO Yanbo. Analysis of the spatial and temporal characteristics of public opinion about Yunnan Tonghai Ms5.0 earthquake in 2018 based on Sina micro-blog[J]. Journal of Seismological Research, 2018, 41(4):525-533. (in Chinese)
[13] 黎洁仪, 梁之彦, 范绍佳.基于位置微博的大风信息提取及应用[J].气象科技, 2019, 47(5):879-884. LI Jieyi, LIANG Zhiyan, FAN Shaojia. Information extraction of strong winds and application based on location microblogs[J]. Meteorological Science and Technology, 2019, 47(5):879-884. (in Chinese)
[14] 邬群勇, 裘钰娇.微博数据位置信息反映台风灾情的有效性分析[J].测绘科学技术学报, 2019, 36(4):406-411. WU Qunyong, QIU Yujiao. Effectiveness analysis of typhoon disaster reflected by microblog data location information[J]. Journal of Geomatics Science and Technology, 2019, 36(4):406-411. (in Chinese)
[15] Wang Y, Wang T, Ye X, et al. Using social media for emergency response and urban sustainability:A case study of the 2012 Beijing rainstorm[J]. Sustainability, 2016, 8(1):25.
[16] 百度地图开放平台[DB/OL].http://lbsyun.baidu.com,2020-04-06. Baidu Map Open Platform[DB/OL].http://lbsyun.baidu.com,2020-04-06.(in Chinese)
[17] 高德开放平台[DB/OL].https://lbs.amap.com,2020-04-06. Gaode Map Open Platform[DB/OL].https://lbs.amap.com,2020-04-06.(in Chinese)
[18] 搜狗输入法.上海地名街道名[DB/OL].https://pinyin.sogou.com/dict/detail/index/9211,2020-03-15. Sogou Input Method. Shanghai place names and street names[DB/OL].https://pinyin.sogou.com/dict/detail/index/9211,2020-03-15.(in Chinese)
[19] Blei D M, Ng A Y, Jordan M I. Latent dirichlet allocation[J]. The Journal of Machine Learning Research, 2003, 4(3):993-1022.
[20] 苏凯, 程昌秀, Nikita Murzintcev, 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, Nikita Murzintcev, 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. (in Chinese)
[21] 刘丽华.自然灾害微博舆情的社会计算模型建构研究[D].武汉大学, 2018. LIU Lihua. Research on Social Computing Model Construction of Natural Disasters Weibo Public Cpinion-based on Sentiment Analysis and Topic Modeling[D]. Wuhan University, 2018. (in Chinese)
[22] 吴丹露, 魏彤, 许家清.R语言环境下的文本可视化及主题分析-以社会服务平台数据为例[J].宁波工程学院学报, 2015, 27(1):19-25. WU Danlu, WEI tong, XU Jiaqing. Text visualization show and topic model analysis based on R software-A case study of social service platform data[J]. Journal of Ningbo University of Technology, 2015, 27(1):19-25. (in Chinese)
[23] Julia S, David Robinson.文本挖掘-基于R语言的整洁工具[M].北京:机械工业出版社, 2018. Julia S, David Robinson.Text Mining with R:A Tidy Approach[M]. Beijing:China Machine Press, 2018.(in Chinese)

备注/Memo

备注/Memo:
收稿日期:2020-07-24;改回日期:2020-12-26。
基金项目:上海市2020年度"科技创新行动计划"社会发展科技攻关定向项目(20DZ1200403);上海市气象局业务类科研项目(MS202008)共同资助
作者简介:杨辰(1988-),男,工程师,硕士,主要从事气象灾害风险分析与研究.E-mail:yangc@lightning.sh.cn
更新日期/Last Update: 1900-01-01