计算机应用研究2016，Vol.33Issue(12)：3676-3680,5.DOI:10.3969/j.issn.1001-3695.2016.12.035
融合主题与语言模型的蒙古文信息检索方法研究
Mongolian information retrieval method based on topic model and language model
摘要
Abstract
Aiming at the retrieval semantic information in Mongolian,this paper proposed a new method combined topic model latent dirichlet allocation(LDA)and language model.This method modeled Mongolian documents with LDA and language mo-del,estimated parameters with Gibbs sampling and represented probability of word,it could mine the hidden relationship be-tween the different topics and the words from documents,got the topic distribution and computed the similarity of keywords top-ics.Finally,it returned to the most relevant documents with topics.Experimental results show that the method has a higher per-formance in topic semantic compared with one sole model.关键词
蒙古文/语言模型/主题模型/Gibbs采样/信息检索Key words
Mongolian/language model/topic model/Gibbs sampling/information retrieval分类
信息技术与安全科学引用本文复制引用
斯日古楞,林民,田长波..融合主题与语言模型的蒙古文信息检索方法研究[J].计算机应用研究,2016,33(12):3676-3680,5.基金项目
国家自然科学基金资助项目（61562068）；内蒙古自然科学基金资助项目（2013MS0912）；内蒙古自治区教育部人文社会科学研究项目 （）
