Research on Sentiment Analysis of Network Forum Based on BP Neural Network
- PDF / 1,153,868 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 98 Downloads / 218 Views
Research on Sentiment Analysis of Network Forum Based on BP Neural Network Yushou Tang 1 & Jianhuan Su 1 & Muazzam A. Khan 2 Accepted: 19 November 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Nowadays, people pay more and more emotional to the emotional analysis of specific goals. Due to the long training time of many networks, this paper proposes a neural network with specific Objective sentiment analysis. Compared with the current neural network, the algorithm proposed in this paper has a shorter training time, which can effectively make up for the lack of emotional mechanism. Finally, we use the emotional data set to carry out simulation experiments. The experimental results show that the proposed algorithm is better than the ordinary neural network algorithm. Keywords Emotional analysis . Neural network . Algorithm
1 Introduction With the popularity of computers, people begin to express their feelings through the Internet. We can extract these data and then classify the data so that we can perceive what opinions and emotions people on the Internet have about what things. With consistent advancement in the field of research, deep learning has become very prominent and researchers have started making sentiment analysis of social network through deep learning. Ref [1] proposed sentiment analysis through recurrent variants latterly on the convolutional neural network of Twitter. Ref [2] proposed facial expression recognition is based on fusion features of center-symmetric local signal magnitude pattern. Ref [3] proposed an experimental study of speech emotion recognition based on deep convolutional neural networks. These manipulations of deep neural
* Jianhuan Su [email protected] Yushou Tang [email protected]
networks incorporate better classification effects than before as for sentiment analysis tasks. Xu [4] et al. in 2018 executed emotional mechanisms in image classification tasks for the very first time and verified the effectiveness of emotional mechanisms in image processing fields. They combined neural networks with the emotional mechanism. Sadr worked out machine translating tasks by combining the emotional mechanism with a re-currentneural network and successfully integrated the emotional mechanism was into the domain of natural language processing. With continuous progression in this research area, Ref [5] proposed a BP based on the emotional mechanism in 2018 and applied it in sentence pair modeling tasks. Wang et al. performed a sentence relation classification using BP based on multi-layer emotional mechanisms. Both the proposed methods were very effective in combination with emotional mechanisms and BP.
2 Problem descriptions In this section, we provide a detailed literature review of relevant studies about our research.
Muazzam A. Khan [email protected]
2.1 Absa
1
Hechi University, RD42, Longjiang Yizhou, Hechi 546300, Guangxi, China
2
Department of Computer Science, Quaid-i-Azam University, Islamabad, Pakistan
ABSA is the sentiment analysis
Data Loading...
