摘要
随着民族地区信息化建设的不断推进,中国少数民族语言网络舆情研究也逐渐引起了大家的关注,文本分类和情感分析模块是舆情系统的重要组成部分。传统的文本分类方法主要通过统计字面上的词语重复次数,而对于文字背后的语义关联考虑甚少。该文重点介绍了一种基于LDA模型在少数民族语言(以彝文为例)网络舆情信息情感分析方面的应用,对文字隐含的主题进行建模,通过挖掘少数民族网页上的舆情信息所蕴含的主题,以及对这些主题进行情感分析,在事件全面爆发之前,采取应急措施。
With the development of ethnic areas of information technology, the Chinese minority language network public opinion research has gradually attracted everyone's attention, text classification and sentiment analysis module is an important part of public opinion of the system. Traditional text classification methods, mainly through word repetitions statistics literally, and semantic association little consideration for the text behind. This article focuses on the LDA model based on minority languages (with Yi for example) the application of information network public opinion sentiment analysis aspects of the theme of the text implied modeling, data mining minorities through public opinion on a web page that contains the theme, as well as sentiment analysis of these topics, before the incident broke out, Bian take emergency measures.
出处
《科技创新导报》
2014年第30期185-186,189,共3页
Science and Technology Innovation Herald
基金
国家自然科学基金"云南跨境民族网络舆情信息挖掘关键技术研究"(项目批准号61363085)项目
云南省重大项目"云南跨境民族语言网络敏感信息传播与分析-以彝文为例"项目(项目批准号ZD2013013)
云南省社会科学项目-云南跨境民族语言网络敏感信息传播与分析(项目批准号YB201152)
云南民族大学创新团队
云南省云南民族大学少数民族语言信息化处理研究中心资助
关键词
主题模型
网络舆情系统
情感分析
Topic model
network public opinion
the detection system