基于LS-SO算法的情感文本分类方法

首先, 基于点互信息与信息检索(PMI IR)算法, 提出一种Laplace平滑情感判定(LS-SO)算法, 对情感词典与表情符号情感词典进行自动扩充, 得到了具有一定规模、 高质量的情感词典, 包括基础情感词典、 目标情感词典、 网络用语情感词典、 表情符号情感词典、 否定词词典、 疑问词词典、 程度副词词典和连词词典. 其次, 通过细化文本语义分析规则计算文本情感值. 实验结果验证了该方法的有效性.

Emotional Text Classification Method Based on LS-SO Algorithm

Firstly, based on the pointwise mutual information information retrieval (PMI IR) algorithm, we proposed Laplace smoothing sentiment opinion (LS SO) algorithm, which automatically expanded theemotional dictionary and the emotional dictionary of emotive symbols. A certainscale and high quality emotional dictionary was obtained, including basic emotional dictionary, target emotional dictionary, online emotional dictionary,emotional dictionary of emotive symbols, negative dictionary, interrogative dictionary, degree adverb dictionary, conjunction dictionary. Then,the emotional value of text was calculated by refining the rules of text semantic analysis. The experimental results verified the effectiveness of the proposed method.