|Table of Contents|

Microblog Emotion Analysis Based on Deep Learning and Attention Mechanism(PDF)

《南京师大学报（自然科学版）》[ISSN:1001-4616/CN:32-1239/N]

Issue:
2023年02期
Page:
115-121
Research Field:
计算机科学与技术
Publishing date:

Info

Title:
Microblog Emotion Analysis Based on Deep Learning and Attention Mechanism
Author(s):
Zhou Xiangzhen1; 2; Li Shuai2; Sui Dong3
(1.School of Information Engineering,Zhengzhou Shengda College of Economics and Trade Management,Zhengzhou 451191,China)
(2.School of Computer Science,Beijing University of Aeronautics and Astronautics,Beijing 100191,China)
(3.School of Electrical and Information Engineering,Beijing University of Civil Engineering and Architecture,Beijing 102406,China)
Keywords:
microblog emotion;  deep learning;  recurrent neural network;  attention mechanism
PACS:
TP312; G254
DOI:
10.3969/j.issn.1001-4616.2023.02.015
Abstract:
In order to improve the performance of Weibo's sentiment analysis, the recurrent neural network in the deep learning algorithm is used for sentiment classification, and the attention mechanism is used to select and weight the word features, so as to enhance the classification accuracy of the recurrent neural network. First, the Weibo corpus is denoised, segmented and vectorized to form an initial sample of Weibo. Then, the Weibo classification model of recurrent neural network is constructed, and the word feature vector is obtained through the node circulation of hidden layer and the output of hidden layer at historical moment and current moment. Then, the attention mechanism is used to calculate the similarity of word features and select weights to construct sentence features, and the classification results are obtained by Softmax function. Finally, the reliability of the proposed method is verified by the Weibo emotion classification simulation test. Experimental results show that, compared with the commonly used Weibo emotion classification algorithm, the proposed method shows higher classification performance under different word vector size samples by setting the attention mechanism window size reasonably.

References:

[1]马晓慧,马尚才,闫俊伢,等. 基于距离感知的目标情感分类模型[J]. 南京师大学报(自然科学版),2021,44(4):111-116.
[2]段吉东,刘双荣,马坤,等. 基于集成学习的文本情感分类方法[J]. 济南大学学报(自然科学版),2019,33(6):483-488.
[3]朱亚军,次曲,拥措. 基于SVM算法的藏文微博情感分析研究[J]. 计算机仿真,2022,39(8):226-229.
[4]冯媛媛,刘克剑,李伟豪. 基于BiLSTM+Self-Attention的多性格微博情感分类[J]. 西华大学学报(自然科学版),2022,41(1):67-76.
[5]NIU Z,ZHONG G,YU H. A review on the attention mechanism of deep learning[J]. Neurocomputing,2021,452:48-62.
[6]GUO M H,XU T X,LIU J J,et al. Attention mechanisms in computer vision:a survey[J]. Computational visual media,2022,8(3):331-368.
[7]张旭辉,张郴,李雅南,等. 城市旅游餐饮体验的注意力机制模型建构——基于机器学习的网络文本深度挖掘[J]. 南京师大学报(自然科学版),2022,45(1):32-39.
[8]LI J,JIN K,ZHOU D,et al. Attention mechanism-based CNN for facial expression recognition[J]. Neurocomputing,2020,411:340-350.
[9]WANG Y,WU H,ZHANG J,et al. Predrnn:a recurrent neural network for spatiotemporal predictive learning[J]. IEEE transactions on pattern analysis and machine intelligence,2022,45(2):2208-2225.
[10]HEWAMALAGE H,BERGMEIR C,BANDARA K. Recurrent neural networks for time series forecasting:current status and future directions[J]. International journal of forecasting,2021,37(1):388-427.
[11]YIN B,CORRADI F,BOHTÉ S M. Accurate and efficient time-domain classification with adaptive spiking recurrent neural networks[J]. Nature machine intelligence,2021,3(10):905-913.
[12]KISVARI A,LIN Z,LIU X. Wind power forecasting—a data-driven method along with gated recurrent neural network[J]. Renewable energy,2021,163:1895-1909.
[13]WEI X,ZHANG L,YANG H Q,et al. Machine learning for pore-water pressure time-series prediction:application of recurrent neural networks[J]. Geoscience frontiers,2021,12(1):453-467.
[14]邓君,孙绍丹,王阮,等. 基于Word2Vec和SVM的微博舆情情感演化分析[J]. 情报理论与实践,2020,43(8):112-119.
[15]吴仁彪,乔晗,贾云飞,等. 基于胶囊网络的中长微博情感分析[J]. 信号处理,2022,38(8):1632-1641.
[16]林伟. 基于PSO-LSTM的中文微博情感分类研究[J]. 中国人民公安大学学报(自然科学版),2022,28(1):95-101.

Memo

Memo:
-
Last Update: 2023-06-15