计算机应用与软件2016，Vol.33Issue(8)：18-22,109,6.DOI:10.3969/j.issn.1000-386x.2016.08.004
微博中的开放域事件抽取
EXTRACTING OPEN DOMAIN EVENTS IN MICROBLOGS
摘要
Abstract
With the rapid development of Internet,the extraction of network information events has been the focus of the study.We thoroughly studied the extraction issue of open domain events in microblogs,and implemented a system of event extraction and categorisation. We characterised the corresponding events by the named entities and event-referring phrases in microblogging sentences mainly extracted with sequence-labelling method,and used the unsupervised categorisation method to classify events.After sorting the events of various categories in every date according to their significances,we displayed them in the form of calendar.In it,we used the conditional random fields to complete the sequence labelling tasks of the event extraction,for unsupervised method we chose the LDA topic model.Experiments prove that the method is effective and feasible.Both the named entity recognition and event-referring phrases extraction achieve high accuracy and recall rates.关键词
事件抽取/条件随机场/文本分类/LDA 模型Key words
Event extraction/Conditional random fields/Text categorisation/Latent Dirichlet allocation (LDA)model分类
信息技术与安全科学引用本文复制引用
陈箫箫,刘波..微博中的开放域事件抽取[J].计算机应用与软件,2016,33(8):18-22,109,6.基金项目
国家自然科学基金项目（61005001）。 （）
