基于英文博客文本的情感分析研究

摘要

随着互联网的迅猛发展,对网络文本带有观点性内容的主观性信息的自动情感分析成为了近期Web信息处理的一个新兴研究热点.文中对英文博客文本的情感分析进行了研究.提出了以词为单位进行特征的选取,并在文本的情感分析中加入了预处理过程.使用机器学习的方法,采用电影评审文本作为语料库,选取在英文文本分类领域较好的支持向量机作为训练器,特征的选取分别采用信息增益、互信息和x2统计的方法.实验结果表明,该方法获得了较满意的结果,与传统的基于N-Gram特征选取方法及基于无监督学习的方法相比,该方法较之都有所提高.%With the rapid development of Internet, the automatic sentiment analysis for subjectivity text become recent research hotspots. Study the sentiment analysis for English weblog text. It used the word as the unit for feature selection and add the pretreatment in text sentiment analysis process. Applied the machine learning method to movie review text sentiment analysis, selected the support vector machine to construct the classifier, used three feature selection methods; information gain, mutual information and chi-square statistic. The results indicate that obtain satisfactory effect, compared with the traditional method based N-Gram and the unsupervised learning approaches, its results have improved.