社会文本流数据富含上下文环境信息、语言不规范且参与用户数量庞大。针对这类数据开展爆发主题探测需要寻找新的思路。本文对社会文本流数据的概念、特点以及爆发主题表达形式进行系统性梳理,从文本内容、时间、社会三个维度阐述探测爆发主题的主要研究思路和基本流程,分析利用社会特征(如用户参与、上下文环境、社团结构)进行爆发主题探测的主要技术方法。 Social text flow data is rich in context information, non-standard language and a large number of participating users. In order to launch the thematic exploration of such data, we must find new ideas. This paper systematically sorts out the concept and characteristics of social text flow data as well as the outbreak of the theme expression form, elaborates the main research ideas and basic flow of the detection explosion theme from the three dimensions of text content, time and society, analyzes the use of social features (such as user participation , Context, community structure) the main technical methods of detection of the outbreak of the theme.