講演抄録/キーワード |
講演名 |
2021-11-26 16:15
Deep Learning Hybrid Models for Sentiment Analysis ○Yunpeng Rong・Jun Ohkubo(Saitama Univ.) NC2021-30 |
抄録 |
(和) |
(まだ登録されていません) |
(英) |
Sentiment analysis (SA), which can analyze the public attitudes towards various texts, has earned increasing attention from researchers. Recent research confirmed that deep learning models show great potential to accomplish the SA tasks because of their flexible structure. This research introduces a hybrid deep learning-based model named attention pooling-based dilated convolution neural network (APDCNN) for implementing the SA tasks. We also conducted experiments to compare its performance with some other previous models, whose results show that dilated operation enhances the performance of our hybrid model without increasing its computational complexity. |
キーワード |
(和) |
/ / / / / / / |
(英) |
Sentiment analysis / Attention pooling / Dilated convolution neural network / / / / / |
文献情報 |
信学技報, vol. 121, no. 271, NC2021-30, pp. 13-17, 2021年11月. |
資料番号 |
NC2021-30 |
発行日 |
2021-11-19 (NC) |
ISSN |
Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
PDFダウンロード |
NC2021-30 |