A document-level model for tweet event detection

Social media like Twitter who serves as a novel news medium and has become increasingly popular since its establishment. Large scale first-hand user-generated tweets motivate automatic event detection on Twitter. Previous unsupervised approaches detected events by clustering words. These methods detect events using burstiness,which measures surging frequencies of words at certain time windows. However,event clusters represented by a set of individual words are difficult to understand. This issue is addressed by building a document-level event detection model that directly calculates the burstiness of tweets,leveraging distributed word representations for modeling semantic information,thereby avoiding sparsity. Results show that the document-level model not only offers event summaries that are directly human-readable,but also gives significantly improved accuracies compared to previous methods on unsupervised tweet event detection,which are based on words/segments.

基于双层注意力机制的篇章级事件真实性检测

自然语言文本中的事件真实性指作者对给定事件存在于客观世界中的确定性程度的描述,正确识别文本中事件的真实性,有助于对自然语言进行深层语义理解。同时,事件真实性检测对诸多自然语言处理应用,如观点检测、事件图谱构建、情感分析等具有重要意义。目前,大多数事件真实性检测研究集中在句子级任务上,而在同一篇章中,经常出现针对同一事件真实性表述不同的情况,此时仅在句子层面识别事件真实性可能会导致矛盾。针对该问题,该文提出了一个基于双层注意力机制的篇章级事件真实性检测方法。首先,利用预训练语言模型BERT对句子进行编码;其次,采用图注意力网络学习句子中的上下文信息与事件之间的依赖关系;最后,利用文档级注意力机制抽取不同句子序列之间的潜在关联,从事件序列集合中获取篇章级事件真实性的最终特征表示。实验结果验证了该方法的有效性,在英文和中文数据集上的实验结果显示,该文所提出方法F1值分别达到87.91%和87.92%,与目前最好系统相比,分别提升了1.40%和1.28%。