Research on the Relationship between Emergent Event Extraction and Evolution—The Case of the Emergency Services Platform
Research on the extraction and evolutionary relationship of emergencies can help quickly and accurately under-stand event situations,reveal complex event relationships,analyze event development trends,predict possible consequenc-es,and provide a basis for scientific emergency service decision-making.In this paper,we crawled emergency incident da-ta from the emergency service network.We utilized natural language processing technology and deep learning algorithms to extract structured information,correlations,and evolutionary relationships of events from unstructured text.We con-structed a comprehensive process for event extraction and evolutionary relationships known as"scenario elements-event"for emergencies,so as to clarify the causes,correlation information,and potential consequences of emergencies.The study also clarifies causes and related information of emergencies,details the evolution process of emergencies,and validates the method on the collected dataset.We constructed an emergency event recognition and classification model,an event entity extraction model,an event relationship extraction model,and an event evolution relationship model.Operating effects of different models are compared,and experimental results confirm the effectiveness of the proposed models in this study,en-riching existing methods for emergencies.
万方数据
