【摘要】 事件抽取是构建事理图谱的重要环节。近年来，由于深度学习的不断发展，对事件抽取的研究产生了重要的影响，利用深度学习技术进行事件抽取已然成为当前主流的事件抽取方法。该文对当前的事件抽取方法进行归纳总结，囊括了融合深度学习方法之后的最新研究成果，以期为该领域的深入研究提供参考。首先，简要叙述事件抽取的主要任务和效果评测指标。接着，对现有的两种事件抽取方法，即基于模板匹配的方法、基于机器学习的方法(基于浅层机器学习和基于深度学习),进行了详细介绍。最后，总结事件抽取现阶段的挑战以及未来的发展趋势。研究表明：随着深度学习的蓬勃发展，事件抽取存在的技术难题不断得到解决，将深度学习技术应用到事件抽取任务以提升抽取性能已是大势所趋。更多还原

【Abstract】 Event extraction is an important part of building an event graph. In recent years, due to the continuous development of deep learning, the research on event extraction has had an important impact. Event extraction using deep learning technology has become the current mainstream event extraction method. We summarize the current event extraction methods, including the latest research results after integrating deep learning methods, in order to provide a reference for in-depth research in this field. Firstly, we briefly describe the main tasks and effect evaluation indicators of event extraction. Then, the two existing event extraction methods are introduced in detail, namely the method based on template matching and the method based on machine learning(based on shallow machine learning and based on deep learning). Finally, we summarize the current challenges of event extraction and future development trends. Research shows that with the vigorous development of deep learning, the technical problems of event extraction are constantly being solved, and it is an irresistible trend to apply deep learning technology to event extraction tasks to improve extraction performance.更多还原