Sentiment Analysis Across Languages: Evaluation Before and After Machine Translation to English

TL;DR

This paper evaluates how transformer models perform on sentiment analysis tasks across multiple languages, both in original form and after machine translation to English, highlighting challenges and future directions.

Contribution

It provides a comparative analysis of transformer-based sentiment analysis performance across languages and translation states, revealing language-specific challenges and insights.

Findings

Transformer models show varying accuracy across languages.

Machine translation can improve or hinder sentiment analysis performance.

Performance disparities highlight the need for language-specific models.

Abstract

People communicate in more than 7,000 languages around the world, with around 780 languages spoken in India alone. Despite this linguistic diversity, research on Sentiment Analysis has predominantly focused on English text data, resulting in a disproportionate availability of sentiment resources for English. This paper examines the performance of transformer models in Sentiment Analysis tasks across multilingual datasets and text that has undergone machine translation. By comparing the effectiveness of these models in different linguistic contexts, we gain insights into their performance variations and potential implications for sentiment analysis across diverse languages. We also discuss the shortcomings and potential for future work towards the end.