：科学交流研究，是情报学长期关注的研究主题，其中非正式科学交流尤其是线下的交流活动作为一种重要的信息交流方式，但长期以来由于受到数据的限制一直不能进行大规模的研究。文章通过探索学术新闻中对于非正式科学交流活动的记录，将活动拆解为参与者、时间、地点、事件类型、报告题目、观点等元素，并对这些元素在新闻中的文本特征做了分析。研究证明利用信息新闻抽取，可以实现对非正式科学交流事件的了解，并可以以此为基础研究线下非正式科学交流的某些特征和规律。

Scientific communication is a hot research topic in information science. Informal communication, especially those happened in real life is an important type of communication. However, there is no large scale study on this topic due to a lack of data. In this study we explore how these activities has been described in academic news and split a piece of news into participants, time, place, event, report title and opinion. We studied the patterns of how these elements are phrased. Based on these knowledge we use a program to implement a process which enables us to extract 160,312 communication activities from news crawled using Bing API and an expert list. Then we carried out a general analysis and a case study.