Abstract
Existing text sentiment analysis methods mostly rely on a large number of language knowledge and sentiment resources. This paper proposes the Multi-channel convolution and bidirectional GRU multi-head attention capsule (AT-MC-BiGRU-Capsule), which uses vector neurons to replace scalar neurons to model text emotions, and uses capsules to characterize text emotions. In addition, traditional methods cannot extract the multi-level features of text sequence well. Multi-head attention can encode the dependencies between words, capture sentiment words in text, and using Convolutional Neural Network (CNN) and Bidirectional gated recurrent unit network (Bi-GRU) to extract local features and global semantic features of text respectively, the global average pooling layer is introduced to obtain the multi-level feature representation of the text sequence more comprehensively. This paper selects three English datasets and one Chinese dataset in the general corpus of sentiment classification to conduct experiments, and achieves better results than other baseline models.
Highlights
In recent years, the Internet has evolved from a static one-way information carrier to a dynamic interactive media, in which more and more users publish news or product reviews to express their opinions
The machine learning method ignores the order of words in the sentence and cannot distinguish the semantics of the sentence, it leads to the problem of sentiment classification error [4]
The main contributions of this paper are as follows: (1) Multi-head attention capsule model combining convolutional neural network and bidirectional GRU network is proposed to be applied to text sentiment analysis tasks
Summary
The Internet has evolved from a static one-way information carrier to a dynamic interactive media, in which more and more users publish news or product reviews to express their opinions. This paper adopts the deep learning method, based on the capsule model of literature [12], and proposes a multi-head attention capsule model that combines convolutional neural network and bidirectional gated recurrent unit (Bi-GRU) to solve the problem of text sentiment analysis. The main contributions of this paper are as follows: (1) Multi-head attention capsule model combining convolutional neural network and bidirectional GRU network is proposed to be applied to text sentiment analysis tasks. This model combines the attention mechanism to construct sentiment capsules for each sentiment category, and uses vector neurons (capsules) to perform text sentiment information Feature representation enhances model generalization ability and improves model robustness. Compared with models that need to incorporate language knowledge and emotional information, this model is more concise and has higher classification accuracy. (2) The model integrates the advantages of local feature extraction of convolutional neural networks and the characteristics of Bi-GRU considering contextual semantics, which effectively improves the classification performance of the model. (3) Multi-head attention is introduced into the model to capture emotional words in the text, encode the dependence between words, and improve the feature expression ability of the model
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.
