摘要
近年来,针对互联网在线信息的情感分析已经成为自然语言处理领域的一个研究热点。提出一个基于主题的情感向量空间模型,它将文本的潜在主题特征融入情感模型中,结合情感词典,利用多标签分类算法,对文本中句的情感极性进行分析与研究。实验结果表明,基于主题的情感向量空间模型在句的情感极性判断上取得了令人满意的效果。
The emotion analysis on internet online information has received much attention from natural language processing field in recent years.A novel emotion vector space model based on topics for text sentences was proposed.The new model including the latent topics features,emotion dictionary and multi-label classification algorithm was applied to analyze the polarity of sentences.Experiment result shows that the model is reasonable and effective in recognizing the polarity of sentences.
出处
《计算机科学》
CSCD
北大核心
2014年第3期32-35,共4页
Computer Science
基金
国家自然科学基金(61075056
61273304)
中央高校基本科研业务费专项资金资助
关键词
情感词典
概率主题
多标签分类
情感分析
Emotion dictionary
Probability topics
Multi-label classification
Emotion analysis
