Multi-task Learning for Cross-Lingual Sentiment Analysis
It addresses sentiment analysis for low-resource languages like Croatian by leveraging Slovene data, but it is incremental as it builds on existing BERT models and multi-task learning approaches.
This paper tackled cross-lingual sentiment analysis by using a trilingual BERT-based model to classify Croatian news articles into positive, negative, and neutral sentiments, achieving evaluation in few-shot and zero-shot scenarios.
This paper presents a cross-lingual sentiment analysis of news articles using zero-shot and few-shot learning. The study aims to classify the Croatian news articles with positive, negative, and neutral sentiments using the Slovene dataset. The system is based on a trilingual BERT-based model trained in three languages: English, Slovene, Croatian. The paper analyses different setups using datasets in two languages and proposes a simple multi-task model to perform sentiment classification. The evaluation is performed using the few-shot and zero-shot scenarios in single-task and multi-task experiments for Croatian and Slovene.