基于Word2Vec模型的泥石流多源灾害数据融合研究

打开文本图片集

关键词：泥石流灾害；知识抽取；质量评估；知识融合；Word2Vec

中图分类号：P694文献标志码：A doi：10.3969/j.issn.1000-1379.2025.07.016

引用格式：，，，等.基于Word2Vec模型的泥石流多源灾害数据融合研究[J].人民黄河，2025，47（7）：97-102.

Research on Multi-Source Debris Flow Disaster Data Fusion Based on Word2Vec Model

JIN Lei¹，XU Peng²，LI Jie²，CAI Yingchun1，³，YANG Haibo1，3

（1.School of Water Conservancy and Transportation， Zhengzhou University， Zhengzhou 45ooo1，China; 2.Guiyang Engineering Corporation Limited，Power Construction Corporation of China，Guiyang 550081，China; 3.NationalKeyLaboratoryofTunnel Boring MachineandInteligentOperationand Maintenance，Zhengzhou450oo1，China） Abstract：Underkgdofdeelotoftateodtialteleles，s flowdisastersisilseitactesgiuouedessdaillo tionmodellibrasuchsjbdLteadectustucuredebrsfsastedatabseetlygegaigl intoadatabasetoachevedatafusionBymappingwodsintoaigdimesioalspacetrouhteWordVecodel，texualvocablaryaso vertedintoalvadtoprsentatiosndKeeCAereaplidtoucetimesioalitoftvctorsdthKmeansalgoitmwasusedforlusteringandvisualization.Theresultssowtatindataextractionevaluation，theveragevaluesofosistecy， completenesandacuacyareallabove8Oiheansuaredevaoselow5.CmparingthCAandtSNEdimnsioalitucto metodsusingeeScoSS）taatecseiscSSvue59pefogtd， whichhasanSSueof.3einmostateattCAdesperpefoancditioallhertdeliis strongontealuestadigsoiablefoebsoastedataetaceveragigthOieureftedecod eltoobtaind，taltftallaaoegeiggsdyuli sourcedebrisfasteatadaticsisteissdyduseptchataeactiosioalitutd clusteringuliatelyprovdingabeakouhhcalethoebrisdasterdataogsifatiodmanticocyfu sion，as well as an important technical solution for disaster data integration.

Key words： debris flow disaster;knowledge extraction；quality evaluation； knowledge fusion；Word2Vec

0 引言

受地理与气候条件的综合影响，我国泥石流灾害发生频率较高，是世界上受该类灾害影响较为严重的国家之一[1-5]。（剩余9385字）