CASIA OpenIR  > 数字内容技术与服务研究中心  > 版权智能与文化计算
李科1; 张兴忠2; 冯晓1
Source Publication计算机工程与设计
Other AbstractWhile analysing the sentiment of product reviews,to get more information of reviews and improve the accuracy of sentiment analysis,a new word vector contains multi-features based on weight distributing is
presented .Sentiment lexicon and feature weight algorithm are used to capture the sentiment information and weight information of words,then the word vector based on distributed representation is optimized,and a review is converted to multi-feature matrix.LSTM(long short term memory)networks are used to extract sequence features and dependencies between words of context from multi-features matrix by the ability to capture long-term
dependencies.Finally,the sequence features and dependencies between words of context are applied to sentiment analysis.Experimental results show that the proposed methods achieve higher precision,recall and F-Measure than traditional sentiment analysis based on LSTM network.
Keyword情感分析 情感特征 权重信息 多元特征 词向量
Document Type期刊论文
First Author AffilicationInstitute of Automation, Chinese Academy of Sciences
Recommended Citation
GB/T 7714
李科,张兴忠,冯晓. 基于特征融合和LSTM网络的评论情感分析[J]. 计算机工程与设计,2018(05):15.
APA 李科,张兴忠,&冯晓.(2018).基于特征融合和LSTM网络的评论情感分析.计算机工程与设计(05),15.
MLA 李科,et al."基于特征融合和LSTM网络的评论情感分析".计算机工程与设计 .05(2018):15.
Files in This Item: Download All
File Name/Size DocType Version Access License
基于特征融合和LSTM网络的评论情感分析(412KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[李科]'s Articles
[张兴忠]'s Articles
[冯晓]'s Articles
Baidu academic
Similar articles in Baidu academic
[李科]'s Articles
[张兴忠]'s Articles
[冯晓]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[李科]'s Articles
[张兴忠]'s Articles
[冯晓]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 基于特征融合和LSTM网络的评论情感分析(李科).doc
Format: Microsoft Word
This file does not support browsing at this time
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.