| 摘要: |
| 针对事件间隐藏关系难挖掘、开放域语义歧义多、可解释性差等问题，提出了一种基于本体模型的事件抽取及图谱构建方法。通过本体引导事件抽取，构建得到事件知识图谱，形成包含实体、事件、关系抽取任务的多任务端到端联合抽取框架，实现事件关系推理分析。仿真实验结果表明，该方法能够有效提高事件关系推理的精确率和召回率。 |
| 关键词: 事件关系推理 本体模型 事件抽取 知识图谱 |
| DOI：10.20079/j.issn.1001-893x.250606001 |
|
| 基金项目: |
|
| Event Extraction and Graph Construction Based on Ontology Model |
| LIU Xubo,REN Haiyang,LIU Jingshu,ZHANG Xuejun |
| (1.Unit 91977 of PLA,Beijing 100036,China;2.The 54th Research Institute of China Electronics Technology Group Corporation，Shijiazhuang 050081，China;3.Hebei Key Laboratory of Intelligent Information Perception and Processing,Shijiazhuang 050081,China;4.Faculty of Robot Science and Engineering,Northeastern University,Shenyang 110819,China) |
| Abstract: |
| To address the problem of difficult mining of hidden relationships between events,the large amount of semantic ambiguity in the open domain,and the poor interpretability,an event extraction and graph construction method based on ontology model is proposed.Through ontology-guided event extraction,an event knowledge graph is constructed,forming a multi-task end-to-end joint extraction framework that includes entity,event,and relationship extraction tasks,to achieve event relationship reasoning and analysis.Simulation experiment results show that this method can effectively improve the accuracy and recall of event relationship reasoning. |
| Key words: event relationship reasoning ontology model event extraction knowledge graph |
