摘要
海量的网络媒体信息使得人们在有限的时间内难以全面地掌握一些话题的信息,这样容易导致部分重要信息的遗漏。话题检测与追踪技术正是在这种需求下产生的。这种技术可以从庞大的信息集合中快速准确地获取人们感兴趣的内容。近几年,话题检测与追踪技术已成为自然语言处理领域热门的研究方向,它能把大量的信息有效地组织起来,并使用相关技术从中挖掘出有用的信息,用简洁有效的方式让人们了解一个事件或现象中所有细节以及它们之间的相关性。对话题跟踪的研究背景、相关概念、评测方法以及相关技术进行了综述,并总结了当前的相关技术。
Massive network media information makes it difficult to achieve a comprehenstve unclerstanalng on some topics. We will definitely miss a lot of information in the case of limited time and equipment, so how to get interested information from vast amounts of information quickly and accurately becomes a necessity, topic detection and tracking technology is studied in such demand. In recent years, topic detection and tracking technology has become a popular research direction in the natural language processing field, it can bring together and organize scattered information, understand all the details of a topic or a phenomenon, as well as the correlation between the events in the them from the whole. The paper intro- duces the background and history of the development of technology of topic tracking, and related concept, it also describes systemically the methods adopted by the current systems of topic detection and tracking, and makes a conclusion on the related technology.
出处
《软件导刊》
2013年第4期147-149,共3页
Software Guide
关键词
话题追踪技术
研究综述
语言模型
Topic Detection and Tracking
Vector Space Model
Language Model
作者简介
王卫姣（1987-），女，四川大学计算机学院硕士研究生，研究方向为数据挖掘与自然语言处理。
