Dou Hu, Lingwei Wei, Yaxin Liu, Wei Zhou, Songlin Hu
This paper describes our system designed for SemEval-2023 Task 12: Sentiment analysis for African languages. The challenge faced by this task is the scarcity of labeled data and linguistic resources in low-resource settings. To alleviate these, we propose a generalized multilingual system SACL-XLMR for sentiment analysis on low-resource languages. Specifically, we design a lexicon-based multilingual BERT to facilitate language adaptation and sentiment-aware representation learning. Besides, we apply a supervised adversarial contrastive learning technique to learn sentiment-spread structured representations and enhance model generalization. Our system achieved competitive results, largely outperforming baselines on both multilingual and zero-shot sentiment classification subtasks. Notably, the system obtained the 1st rank on the zero-shot classification subtask in the official ranking. Extensive experiments demonstrate the effectiveness of our system.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Zero-shot Sentiment Classification | AfriSenti | weighted-F1 score | 0.589 | SACL-XLMR |
| Zero-shot Sentiment Classification | AfriSenti | weighted-F1 score | 0.561 | AfroXLMR |
| Zero-shot Sentiment Classification | AfriSenti | weighted-F1 score | 0.439 | AfriBERTa |
| Zero-shot Sentiment Classification | AfriSenti | weighted-F1 score | 0.399 | XLM-R |
| Zero-shot Sentiment Classification | AfriSenti | weighted-F1 score | 0.34 | Random |