Abstract
In view of the limited text features of short texts, features of short texts should be mined from various angles, and multiple sentiment feature combinations should be used to learn the hidden sentiment information. A novel sentiment analysis model based on multi-channel convolutional neural network with multi-head attention mechanism (MCNN-MA) is proposed. This model combines word features with part of speech features, position features and dependency syntax features separately to form three new combined features, and inputs them into the multi-channel convolutional neural network, as well as integrates the multi-head attention mechanism to more fully learn the sentiment information in the text. Finally, experiments are carried out on two Chinese short text data sets. The experimental results show that the MCNN-MA model has a higher classification accuracy and a relatively low training time cost compared with other baseline models.
Highlights
Text sentiment analysis is the process of analyzing, processing, summarizing and judging the sentiment tendency of subjective texts with sentimental colors [1]
In response to the above problems, this paper proposes a text sentiment analysis method that is based on multi-channel convolutional neural network with multi-head
The classification accuracy of the Convolutional Neural Network (CNN)-multi-channel model on the two Chinese data sets is improved by 1.93% and 1.91% respectively than the best model among the CNN model, DCNN model and WFCNN model
Summary
Text sentiment analysis is the process of analyzing, processing, summarizing and judging the sentiment tendency of subjective texts with sentimental colors [1]. Chen et al [4] used the sentiment dictionary to extract the binary features of the words in the text and added the binary features to a convolutional neural network, which achieved better performance on the COAE2014 data set than traditional machine learning methods and basic convolutional neural networks He et al [17] proposed a multi-channel convolutional neural network (EMCNN) with sentiment semantic enhancement. Zhao and Wu [15] proposed a convolutional neural network structure combined with an attention mechanism, adding an attention layer between the input layer and the convolutional layer to create a context vector for each word, and concatenating it with the word vector to form a new vector and send the new vector into convolutional neural network This model can capture long-distance contextual information and the connection between discontinuous words through the attention mechanism, and has a classification effect better than traditional CNN. By concatenating the dependency syntax feature vectors of n words, the dependency syntax feature vector matrix Ps corresponding to a sentence of length n can be obtained, as shown in formula (5)
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
