Downloads: 2 | Views: 116 | Weekly Hits: ⮙1 | Monthly Hits: ⮙2

Research Paper | Neural Networks | Egypt | Volume 12 Issue 2, February 2023

A Russia-Ukraine Conflict Tweets Sentiment Analysis Using Bidirectional LSTM Network

Eman Sedqy Ibraihm Shlkamy | Khaled Mohammed Mahar | Ahmed Ahmed Hesham Sedky

Abstract: Sentiment analysis techniques have a vital role in analyzing people?s opinions. The continuous and rapid growth of data posted on social media sites is the fuel that draws people's opinions. Despite the fact that the vast majority of research focuses on analyzing sentiment to study the impact of the war on the global economy. Furthermore, the actions of national leaders or other powerful figures have typically received more attention in the research of international conflict than public emotions and opinions. The purpose of this paper is to go over some of the most relevant works on sentiment analysis, which are limited by a simple architecture and focus on analyzing public emotions and opinions during the Russia-Ukraine Conflict as compared to leaders and other powerful figures. This paper proposes to use a single bidirectional LSTM network for English tweet sentiment analysis using a classification of positive, negative, and neutral as a multi-class classification approach. We utilized one bidirectional long short-term memory (Bi-LSTM) layer along with the global Max pooling ID mechanism and achieved an accuracy of 91.79%. The results of the proposed framework show well performance over previous studies with complex structures that have previously been proposed. Measuring performance in terms of accuracy. During crises, it is crucial to pay attention to simple architectural models to solve similar problems without complex architectural neural networks.

Keywords: Sentiment Analysis, Bidirectional, LSTM, Deep learning, Natural language processing, NLP

Edition: Volume 12 Issue 2, February 2023,

Pages: 522 - 530