结合话题相关性的热点话题情感倾向研究

发布时间：2018-07-03 20:10

本文选题：热点话题 + 主客观分类　；参考：《数据分析与知识发现》2017年03期

【摘要】：【目的】热点话题具有很大的影响力,针对热点话题及其情感对象的情感倾向进行相关研究。【方法】提出一个结合话题相关性的主客观分类模型,帮助抽取与热点话题相关的主观微博;利用基于机器学习改进的情感分类方法对抽取博文的情感极性进行分析;通过召回率、准确率、F值对情感分类效果进行详细评估。【结果】实证分析结果表明,结合话题相关性有效提升了热点话题微博主客观分类和情感极性分类效果,其中F值分别提升7.4%和2.2%。【局限】待需深入考虑数据的分布状态、情感分类粒度细化、情感对象的情感趋势变化等。【结论】考虑话题相关性,提升微博情感分类的效果,并通过抽取热点话题中关键情感对象的情感倾向,为微博精准营销提供相关情报信息。
[Abstract]:[objective] Hot topic has great influence. This paper studies the emotional tendency of hot topic and its affective object. [methods] A subjective and objective classification model combining topic correlation is proposed. Help to extract subjective Weibo related to hot topics; use improved emotion classification method based on machine learning to analyze emotional polarity of extracted blog posts; recall rate, [results] the empirical results show that the combination of topic correlation can effectively improve the effectiveness of Weibo subjective and objective classification and affective polarity classification. The F value was increased by 7.4% and 2.2% respectively. [limitation] the distribution of data, the granularity of emotion classification, the change of emotional trend of emotional objects should be considered deeply. [conclusion] considering the correlation of topic, the effect of Weibo emotional classification should be improved. And by extracting the emotional tendency of the key emotional objects in the hot topic, this paper provides the relevant information for Weibo precise marketing.
【作者单位】：四川大学商学院;
【基金】：四川大学中央高校基本科研业务费项目“基于中文微博的负面情绪预警研究”(项目编号:skqy201406)的研究成果之一
【分类号】：TP391.1

【相似文献】