Attention-Based Convolution Bidirectional Recurrent Neural Network for Sentiment Analysis

Soubraylu Sivakumar, Haritha D. (https://orcid.org/0000-0003-2772-2081) (ea63803d-6ff6-4ae7-9d66-89474f17d43f, Sree Ram N. (http://orcid.org/0000-0002-2721-7678) (83d6d495-4435-4605-8ac2-1ea7a0deb94f, Naveen Kumar, Rama Krishna G. (https://orcid.org/0000-0003-3572-6517) (a4bc1e39-7e05-4f67-8fb5-014c74b96deb, Dinesh Kumar A. (https://orcid.org/0000-0003-2008-6828) (7d795588-8258-4ca5-8ef2-9e5d29cad026

Source Title: International Journal of Decision Support System Technology (IJDSST)14(1)

ISSN: 1941-6296|EISSN: 1941-630X|EISBN13: 9781683180890|DOI: 10.4018/IJDSST.300368

Cite Article Cite Article

MLA

Sivakumar, Soubraylu, et al. "Attention-Based Convolution Bidirectional Recurrent Neural Network for Sentiment Analysis." IJDSST vol.14, no.1 2022: pp.1-21. http://doi.org/10.4018/IJDSST.300368

APA

Sivakumar, S., Haritha D. (https://orcid.org/0000-0003-2772-2081) (ea63803d-6ff6-4ae7-9d66-89474f17d43f, Sree Ram N. (http://orcid.org/0000-0002-2721-7678) (83d6d495-4435-4605-8ac2-1ea7a0deb94f, Kumar, N., Rama Krishna G. (https://orcid.org/0000-0003-3572-6517) (a4bc1e39-7e05-4f67-8fb5-014c74b96deb, & Dinesh Kumar A. (https://orcid.org/0000-0003-2008-6828) (7d795588-8258-4ca5-8ef2-9e5d29cad026. (2022). Attention-Based Convolution Bidirectional Recurrent Neural Network for Sentiment Analysis. International Journal of Decision Support System Technology (IJDSST), 14(1), 1-21. http://doi.org/10.4018/IJDSST.300368

Chicago

Sivakumar, Soubraylu, et al. "Attention-Based Convolution Bidirectional Recurrent Neural Network for Sentiment Analysis," International Journal of Decision Support System Technology (IJDSST) 14, no.1: 1-21. http://doi.org/10.4018/IJDSST.300368

Export Reference

Favorite Full-Issue Download

View Full Text HTML

View Full Text PDF

Abstract

A customer conveys their opinion in natural language about an entity. Applying sentiment analysis to those reviews is a very complex task. The significant terms that influence the polarity of a review are not examined. The terms that have contextual meaning are not recognized and are present across multiple sentences in a review. To address the above two issues, the authors have proposed an attention-based convolution bi-directional recurrent neural network (ACBRNN). In this model, two convolution layer captures phrase-level feature while self-attention in the middle assigns high weight to the significant terms, and bi-directional GRU performs a conceptual scanning of review through forward and backward direction. The authors have conducted four different experiments (i.e., unidirectional, bidirectional, hybrid, and proposed model) on IMDB dataset to show the significance of the proposed model. The proposed model has obtained an F1 score of 87.94% on IMDB dataset, which is 5.41% higher than CNN. Thus, the proposed architecture performs well compared with all other baseline models.