摘要
现有的话题追踪方法大多面向新闻数据,将其应用于论坛时效果不够理想。结合论坛的特点,提出一种基于语义相似度的论坛话题追踪方法。该方法首先通过构建话题和帖子的关键词表建立其文本表示模型,然后利用知网计算两个关键词表的语义相似度并以此作为帖子与话题的相关程度,最后根据相关程度实现论坛话题追踪。该方法较好地避免了向量空间模型的缺陷。实验表明,该方法能比较有效地解决面向论坛的话题追踪问题。
To study the BBS topic tracking, the paper discovered that most of the traditional methods of topic tracking deal with news reports, and they are not suitable when they are applied to BBS. The paper utilized the characteristics of BBS and presented a topic tracking method for BBS data based on semantic similarity. This method firstly constructed keywords tables of topic and post as their representation models, and then computed the two tables＇ semantic similarity with the help of HowNet which is served as correlation degree between post and topic. Finally, this method used the correlation degree to realize BBS-oriented topic tracking. This method effectively avoids the disadvantage of Vector Space Model （ VSM）. The experimental results show that this method can solve the problem of BBS-oriented topic tracking effectively.
出处
《计算机应用》
CSCD
北大核心
2011年第1期93-96,共4页
journal of Computer Applications
基金
国家863计划项目(2007AA01Z439)
关键词
话题追踪
论坛
关键词
语义相似度
向量空间模型
topic tracking
BBS
key word
semantic similarity
Vector Space Model （VSM）
