作者: Frederik Hogenboom , Flavius Frasincar , Uzay Kaymak , Franciska de Jong , Emiel Caron
DOI: 10.1016/J.DSS.2016.02.006
关键词:
摘要: Event extraction, a specialized stream of information extraction rooted back into the 1980s, has greatly gained in popularity due to advent big data and developments related fields text mining natural language processing. However, up this date, an overview particular field remains elusive. Therefore, we give summarization event techniques for textual data, distinguishing between data-driven, knowledge-driven, hybrid methods, present qualitative evaluation these. Moreover, discuss common decision support applications from corpora. Last, elaborate on systems identify current research issues. We approaches.A wide variety can benefit extraction.Pressing issues be addressed are scalability domain dependencies.Evaluation with annotated standard benchmarks or crowdsourcing is advised.