摘要
在构建一个完整的情感词典的基础上,利用情感词典进行情感特征选择,并考虑了中文微博语料不均衡对情感特征选择的影响,在统计中引入了特征项频度因素进行特征项降维,并根据选择的情感特征项和权重因子,使用SVM分类器对中文微博进行情感分类.实验结果表明,该方法是有效的.
Based on the construction of a complete emotional dictionary,microblog emotional features are selected.In the paper the impact of uneven corpus is taken into account,the characteristic values are extracted by adding the characteristics of frequency factor in the chi-square statistics method,and microblog emotions are analyzed by the SVM classifier according to emotional characteristics and weighting factor.The experimental results show that the method is effective.
出处
《内蒙古师范大学学报(自然科学汉文版)》
CAS
北大核心
2016年第1期84-88,共5页
Journal of Inner Mongolia Normal University(Natural Science Edition)
基金
湖北省教育厅重点科研项目(D20145001)
关键词
中文微博
情感特征
词典
卡方统计
Chinese microblog
sentiment feature
lexicon
chi-square statistics