###
计算机系统应用英文版:2017,26(3):182-186
本文二维码信息
码上扫一扫!
基于词向量模型的情感分析
(1.中国科学院大学, 北京 100049;2.中国科学院计算机网络信息中心, 北京 100190)
Sentiment Analysis Based on Word Vector Model
(1.University of Chinese Academy of Sciences, Beijing 100049, China;2.Computer Network Information Center, Chinese Academy of Sciences, Beijing 100190, China)
摘要
图/表
参考文献
相似文献
本文已被:浏览 1970次   下载 4225
Received:June 21, 2016    Revised:August 18, 2016
中文摘要: 随着移动互联网的发展,以商品评论等带有主观性的短文本信息急剧增加.海量的文本信息使得人工管理越来越困难.本文以商品评论为研究对象进行情感分析.针对商品评论为短文本的特点,本文在词向量的基础上提出了词向量叠加方法和加权词向量方法进行文本特征的提取,从而更深层次的提取短文本特征.在进行评论情感分析模型性能的比较中,说明了本文所提方法的有效性.基于情感分析技术可以解决人工难以胜任的海量商品评论的分类,方便用户快速获取有效信息.
Abstract:With the development of Internet, text information, such as product review, increases rapidly. The mass text information makes it more difficult to make artificial management. Considering that product reviews are short text, this paper comes up with the method of word vector superposition and weighted word vector. In the result of sentiment analysis, the method is proved effective. Emotional analysis technology can solve the difficulty of artificial classification in the mass of product review, and help users to get information quickly.
文章编号:     中图分类号:    文献标志码:
基金项目:
引用文本:
魏广顺,吴开超.基于词向量模型的情感分析.计算机系统应用,2017,26(3):182-186
WEI Guang-Shun,WU Kai-Chao.Sentiment Analysis Based on Word Vector Model.COMPUTER SYSTEMS APPLICATIONS,2017,26(3):182-186