Comparative Approaches to Sentiment Analysis Using Datasets in Major European and Arabic Languages

TL;DR

This paper compares transformer-based models like BERT, mBERT, and XLM-R for multilingual sentiment analysis, highlighting XLM-R's superior adaptability and the importance of fine-tuning for underrepresented languages.

Contribution

It demonstrates XLM-R's effectiveness in morphologically complex languages and emphasizes fine-tuning strategies for better sentiment classification in diverse languages.

Findings

XLM-R achieves over 88% accuracy in complex languages

Fine-tuning significantly improves model performance

XLM-R outperforms other models in multilingual sentiment analysis

Abstract

This study explores transformer-based models such as BERT, mBERT, and XLM-R for multi-lingual sentiment analysis across diverse linguistic structures. Key contributions include the identification of XLM-R superior adaptability in morphologically complex languages, achieving accuracy levels above 88%. The work highlights fine-tuning strategies and emphasizes their significance for improving sentiment classification in underrepresented languages.