Natural Language Processing and Sentiment Analysis on Bangla Social Media Comments on Russia-Ukraine War Using Transformers

被引:8
|
作者
Hasan, Mahmud [1 ]
Islam, Labiba [1 ]
Jahan, Ismat [1 ]
Meem, Sabrina Mannan [1 ]
Rahman, Rashedur M. [1 ]
机构
[1] North South Univ, Dept Elect & Comp Engn, Dhaka 1229, Bangladesh
关键词
Natural language processing; sentiment analysis; transformers; Russia-Ukraine war;
D O I
10.1142/S2196888823500021
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The Bangla Language ranks seventh in the list of most spoken languages with 265 native and non-native speakers around the world and the second Indo-Aryan language after Hindi. However, the growth of research for tasks such as sentiment analysis (SA) in Bangla is relatively low compared to SA in the English language. It is because there are not enough high-quality publically available datasets for training language models for text classification tasks in Bangla. In this paper, we propose a Bangla annotated dataset for sentiment analysis on the ongoing Ukraine-Russia war. The dataset was developed by collecting Bangla comments from various videos of three prominent YouTube TV news channels of Bangladesh covering their report on the ongoing conflict. A total of 10,861 Bangla comments were collected and labeled with three polarity sentiments, namely Neutral, Pro-Ukraine (Positive), and Pro-Russia (Negative). A benchmark classifier was developed by experimenting with several transformer-based language models all pre-trained on unlabeled Bangla corpus. The models were fine-tuned using our procured dataset. Hyperparameter optimization was performed on all 5 transformer language models which include: BanglaBERT, XLM-RoBERTa-base, XLM-RoBERTa-large, Distil-mBERT and mBERT. Each model was evaluated and analyzed using several evaluation metrics which include: F1 score, accuracy, and AIC (Akaike Information Criterion). The best-performing model achieved the highest accuracy of 86% with 0.82 F1 score. Based on accuracy, F1 score and AIC, BanglaBERT outperforms baseline and all the other transformer-based classifiers.
引用
收藏
页码:329 / 356
页数:28
相关论文
共 50 条
  • [1] Social Media Analytics on Russia-Ukraine Cyber War with Natural Language Processing: Perspectives and Challenges
    Sufi, Fahim
    INFORMATION, 2023, 14 (09)
  • [2] A multidimensional analysis of media framing in the Russia-Ukraine war
    Ibrahim, Majd
    Wang, Bang
    Xu, Minghua
    Xu, Han
    JOURNAL OF COMPUTATIONAL SOCIAL SCIENCE, 2025, 8 (02):
  • [3] Sentiment Analysis of Russia-Ukraine Conflict Tweets Using RoBERTa
    Ramos, Leo
    Chang, Oscar
    UNICIENCIA, 2023, 37 (01)
  • [4] Sentiment analysis on social media tweets using dimensionality reduction and natural language processing
    Omuya, Erick Odhiambo
    Okeyo, George
    Kimwele, Michael
    ENGINEERING REPORTS, 2023, 5 (03)
  • [5] Using Natural Language Processing to Explore Social Media Opinions on Food Security: Sentiment Analysis and Topic Modeling Study
    Molenaar, Annika
    Lukose, Dickson
    Brennan, Linda
    Jenkins, Eva L.
    Mccaffrey, Tracy A.
    JOURNAL OF MEDICAL INTERNET RESEARCH, 2024, 26
  • [6] Performance Evaluation of Reddit Comments Using Machine Learning and Natural Language Processing Methods in Sentiment Analysis
    Zhang, Xiaoxia
    Qi, Xiuyuan
    Teng, Zixin
    COMPUTATIONAL AND EXPERIMENTAL SIMULATIONS IN ENGINEERING, ICCES 2024-VOL 2, 2025, 173 : 14 - 24
  • [7] Mobility Pattern Analysis during Russia-Ukraine War Using Twitter Location Data
    Shu, Yupei
    Chen, Xu
    Di, Xuan
    INFORMATION, 2024, 15 (02)
  • [8] Conflict and polarisation on social media caused by the Russia-Ukraine War: The case of Ekşi Sözlük
    Gurocak, Tolga
    CONNECTIST-ISTANBUL UNIVERSITY JOURNAL OF COMMUNICATION SCIENCES, 2023, (65): : 1 - 32
  • [9] Impact of social media-based dance therapy in treating depression symptoms among victims of Russia-Ukraine war
    Ahmad, Jamilah
    Okwuowulu, Charles
    Sanusi, Bernice
    Bello, Samson Adedapo
    Talabi, Felix Olajide
    Udengwu, Ngozi
    Gever, Verlumun Celestine
    HEALTH PROMOTION INTERNATIONAL, 2022, 37 (06)
  • [10] Sentiment Analysis: Automated Evaluation Using Natural Language Processing
    Novak, Michal
    CREATING GLOBAL COMPETITIVE ECONOMIES: 2020 VISION PLANNING & IMPLEMENTATION, VOLS 1-3, 2013, : 973 - 975