《计算机应用研究》|Application Research of Computers

基于性格的微博情感分析模型PLSTM

Personality-based microblog sentiment analysis model PLSTM

免费全文下载 (已被下载 次)  
获取PDF全文
作者 袁婷婷,杨文忠,仲丽君,张志豪,向进勇
机构 新疆大学 信息科学与工程学院,乌鲁木齐 830046
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2020)02-005-0342-05
DOI 10.19734/j.issn.1001-3695.2018.07.0521
摘要 不同性格用户所具有的语言表达方式不尽相同,现有情感分析工作很少考虑到用户性格,针对此问题,提出一种基于性格的微博情感分析模型PLSTM。该模型首先采用性格识别规则将微博文本分为五个性格集合和一个通用集合,其次针对每种性格文本集合分别训练出一个情感分类器,最后对六个基本情感分类器进行融合,得出最终的情感极性。实验结果显示PLSTM方法的<i>F<sub>1</sub></i>值可以达到96.95%,表明PLSTM比起基准情感分析模型在准确率、召回率、<i>F<sub>1</sub></i>值上都有较大提高。
关键词 情感分析; 性格; word2vec; 长短时记忆网络; 分类器融合
基金项目 国家自然科学基金资助项目(U1603115,61262087)
国家"973"计划资助项目(2014CB340500)
国家自然科学基金重点项目(U1435215)
本文URL http://www.arocmag.com/article/01-2020-02-005.html
英文标题 Personality-based microblog sentiment analysis model PLSTM
作者英文名 Yuan Tingting, Yang Wenzhong, Zhong Lijun, Zhang Zhihao, Xiang Jinyong
机构英文名 College of Information Science & Engineering,Xinjiang University,Urumqi 830046,China
英文摘要 Users of different personalities have different language expressions. Existing sentiment analysis work rarely considers the personality of the user. To solve this problem, this paper proposed a micro-blog sentiment analysis model based personality, called PLSTM. The model firstly used the personality recognition rules to divide the microblog text into five personality sets and a universal set, then trained a sentiment classifier for each personality set, and finally integrated six basic sentiment classifiers to obtain the ultimate sentiment polarity. The experimental results show that the <i>F<sub>1</sub></i> value of the PLSTM method can reach 96.95%, which indicates that PLSTM has a higher improvement in accuracy, recall rate and <i>F<sub>1</sub></i> value than the commonly used benchmark sentiment analysis model.
英文关键词 sentiment analysis; personality; word2vec; long and short memory network; classifier fusion
参考文献 查看稿件参考文献
 
收稿日期 2018/7/12
修回日期 2018/8/23
页码 342-346
中图分类号 TP391.1
文献标志码 A