摘要
主题情感混合模型(Reverse-Joint Sentiment/Topic Model;Joint Sentiment/Topic Model)能够有效地同时抽取文档的主题和情感信息,在情感分析领域受到广泛的关注,因为没有考虑整体分布与局部分布的关系,导致分类效果不佳且不稳定.本文同时考虑两个粒度上的情感/主题分布——文档级和局部,提出多粒度的主题情感混合模型(MG-.R-JST;MG-JST).MG-R-JST/MG-JST、在文档级分布和局部分布的共同作用下生成单词的情感/主题;使用吉布斯采样进行模型推理,并给出了推理过程;在MR与MDS数据集上进行实验,实验结果表明本文算法分类效果优于主题情感混合模型,且稳定性更好.
The topic and sentiment unification model (Reverse-Joint Sentiment/Topic Model;Joint Sentiment/Topic Model) can effectively extract information of topic and sentiment simultaneously and receives wide attention in the field of sentiment analy- sis,because it does not consider the relationship between the overall distribution and local distribution, so the classification perfor- mance is not good and stable. This paper proposed the multi-grain topic and sentiment unification model (MG-R-JST;MG-JST) by taking into account both grains on sentiment/topic distributiort---document-level and local-level.MG-R-JST/MG-JST generated the sentiment/topic of words on the effect of the document-level and local-level distribution, we used Gibbs sampling for model infer- ence and showed the process.Experiments on the dataset of MR and MDS demonstrate the effectiveness of the proposed method, and the classification performance is better and more stable than the topic and sentiment unification model.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2015年第9期1875-1880,共6页
Acta Electronica Sinica
基金
国家自然科学基金(No.61170092
No.61133011
No.61103091)
关键词
LDA
主题情感混合模型
情感分析
多粒度
LDA
topic and sentiment unification model
sentiment analysis
multi-grain