动态文本会话抽取技术研究  

Research of Dynamic Text Conversation Extraction

作　　者：张瑛[1] 张娅婷[1] 

机构地区：[1]湖南信息职业技术学院计算机工程系,湖南长沙410200

出　　处：《电视技术》2011年第11期84-87,共4页Video Engineering

摘　　要：对海量的短文数据进行自动分析和挖掘,从中获取有价值的隐含知识已经成为一项迫切的需求。动态文本会话抽取是针对MSN,QQ等动态数据,将属于同一会话但相互交错的不同消息文本组织在一起,属于在线话题发现追踪的新兴领域,在信息检索,文本挖掘和话题检测追踪等方面有着重要应用。首先介绍了文本会话抽取的必要性和重要性,介绍其主要研究内容和结果评测方法;然后对其中多个研究内容提出一个统一研究框架,并对该框架中的关键技术进行了详细分析;最后指出该领域中的关键问题及难点,并对未来研究做出展望。Automatically analyzing and deriving implied value of knowledge for massive short data has become an urgent demand. Text conversation extraction is proposed for such dynamic data like MSN and QQ to detect intertwined messages that belong to the same conversations, which is a new area of online topic detection and tracing, and playing an important role in the application of information retrieval, text mining, topic detection tracking, etc. In this survey, the necessity and importance of dynamic text conversation extraction are firstly introduced. The main contents and evaluation methods are then analyzed, an integrated research framework is proposed, and the key technologies of framework are analyzed. Finally, the key issues, difficulties, and future directions in the field of text conversation are all summarized.

关 键 词：文本会话 会话抽取 统一研究框架 

分 类 号：TN915[电子电信—通信与信息系统] TP311[电子电信—信息与通信工程]