Sentiment analysis leveraging emotions and word embeddings
详细信息 查看全文
- 作者：Maria Giatsogloua ; b ; mgiatsog@csd.auth.gr" class="auth_mail" title="E-mail the corresponding author ; Manolis G. Vozalisb ; mans@uom.gr" class="auth_mail" title="E-mail the corresponding author ; Konstantinos Diamantarasb ; kdiamant@it.teithe.gr" class="auth_mail" title="E-mail the corresponding author ; Athena Vakalia ; avakali@csd.auth.gr" class="auth_mail" title="E-mail the corresponding author ; George Sarigiannidisc ; g.sarigiannidis@msensis.com" class="auth_mail" title="E-mail the corresponding author ; Konstantinos Ch. Chatzisavvas ; c ; kchatz@msensis.com" class="auth_mail" title="E-mail the corresponding author
- 关键词：Multilingual sentiment analysis ; Text analysis ; Machine learning ; Vector representation ; Hybrid vectorization ; Online user reviews
- 刊名：Expert Systems with Applications
- 出版年：2017
- 出版时间：1 March 2017
- 年：2017
- 卷：69
- 期：Complete
- 页码：214-224
- 全文大小：798 K
文摘
- •
A flexible, generic methodology for the sentiment prediction of written documents.
- •
The methodology can be easily customized for any language.
- •
Hybrid approach combining Word2Vec and Bag-of-Words representations.
- •
Applied on online user reviews in both Greek and English languages.
- •
Improved accuracy and efficiency in comparison to existing other approaches.