摘要
近年来,微博热点话题发现已经成为当前网络舆情分析研究的热点.本文针对微博信息的碎片化、口语化等短文本特点,为解决向量空间模型(VSM)文本表示方法存在高维度、稀疏,及同义多义等问题,采用潜在语义分析法对微博信息进行建模,再通过贝叶斯分类算法实现话题发现.并采用J2EE开发包及Eclipse集成开发环境,结合Hibernate,Lucene等技术实现了微博热点话题发现系统,实验表明这种方法是有效的.
The hot topics of micro-blog detecting has become the current research focuses of Internet public opinion information. In order to solve the existing problems of high-dimension, sparse, synonymy and polysemy from the Vector Space Model (VSM) text presentation, the micro-blog information model was developed using LSA for the short texts of the fragment, colloquial micro blog informa- tion,then the topic detection was achieved through the Bayesian classification algorithm. Furthermore, the micro blog topic detecting system was constructed by adopting software developer's kit J2EE, the integrated development environment Eclipse and techniques such as Hibernate and Lucene, and the operation of the system was proved to be effective.
出处
《内蒙古科技大学学报》
CAS
2013年第3期283-286,共4页
Journal of Inner Mongolia University of Science and Technology
关键词
语义分析微博
热点
话题发现
semantic analysis micro blogs
hot topics
topic detection