论文部分内容阅读
微博作为一种表达用户观点、传播热点事件的互联网社交工具,已经成为越来越多人日常生活中必不可少的一部分。因此,进行微博的情感分析也成为人们研究的热点。文章采用了机器学习方法中的条件随机场算法和支持向量机算法,结合微博中的表情字符,对微博的主客观分类与情感倾向性分类进行了研究。对比实验表明,条件随机场算法比支持向量机算法在微博情感分类领域有更高的准确度。
As an Internet social tool that expresses user’s point of view and disseminates hot events, weibo has become an indispensable part of more and more people’s daily life. Therefore, the emotional analysis of Weibo has also become a hot spot for people’s research. This paper uses conditional random field algorithm and support vector machine (SVM) algorithm in machine learning method to study the subjective and objective classification and sentiment orientation of Weibo based on the facial expression characters in Weibo. Contrast experiments show that the conditional random field algorithm has higher accuracy than the support vector machine algorithm in the micro-blog emotion classification field.