This paper proposes an unsupervised method of sentiment classification and applies it to perform sentiment classification on Sina micro-blog. The approach employs emotional images and emotional words as the emotional knowledge to extract pseudo-labeled samples, and uses them to train a classifier for automatically classification on polarities of the miro-blog. Experimental results show that the method achieves a decent performance on sentiment classification for Chinese micro-blog.%通过对新浪微博文本进行情感信息方面的分析与研究,提出一种基于情绪知识的非监督情感分类方法.利用情绪词和表情图片2种情绪知识对大规模微博非标注语料进行筛选并自动标注,用自动标注好的语料作为训练集构建微博情感文本分类器,对微博文本进行情感极性自动分类.实验结果表明,该方法对微博文本的情感极性分类达到较好的效果.
展开▼