Multilingual Sentiment Analysis Based on Fine-Tuned Transformer Architectures

View at publisher Summarise with VIRAT
Authors: Reema Ajmera, Anantika Johari, Abhishek Goyal, Abhishek Purohit, Abhishek Kumar, Jha Anisha Ashok
Cited by: 0
References: 15

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

As multilingual social media continues to expand exponentially, online critiques and user-generated content increase, making traditional sentiment analysis methods inadequate for capturing cross-lingual nuances. This paper presents a detailed analysis of multilingual sentiment analysis using a fine-tuned transformer model, including XLM-Roberta. The study covers dataset preparation, preprocessing pipelines, fine-tuning methods, and cross-lingual testing. The proposed framework leverages shared multilingual embeddings, achieving higher accuracy and improved generalization across multiple languages without relying on translation-based training. Ethical considerations, deployment challenges, and future research directions are also discussed.

Fields of Study & Subjects

Sentiment Analysis and Opinion Mining Topic Modeling Hate Speech and Cyberbullying Detection

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
1589–1592
Publication date
21 January 2026
Language
EN