WESSA at SemEval-2020 Task 9: Code-Mixed Sentiment Analysis using Transformers
This work addresses sentiment analysis for code-mixed text, a domain-specific problem, and is incremental as it applies an existing method to a new dataset.
The paper tackled sentiment analysis for code-mixed social media text by fine-tuning XLM-RoBERTa on monolingual and code-mixed data, achieving a 70.1% average F1-Score on the official test set and later improving to 75.9%.
In this paper, we describe our system submitted for SemEval 2020 Task 9, Sentiment Analysis for Code-Mixed Social Media Text alongside other experiments. Our best performing system is a Transfer Learning-based model that fine-tunes "XLM-RoBERTa", a transformer-based multilingual masked language model, on monolingual English and Spanish data and Spanish-English code-mixed data. Our system outperforms the official task baseline by achieving a 70.1% average F1-Score on the official leaderboard using the test set. For later submissions, our system manages to achieve a 75.9% average F1-Score on the test set using CodaLab username "ahmed0sultan".