kk2018 at SemEval-2020 Task 9: Adversarial Training for Code-Mixing Sentiment Classification
This addresses sentiment analysis for multilingual speakers in code-mixed text, representing an incremental advance in a niche domain.
The paper tackled code-mixing sentiment classification by testing domain transfer learning from a uni-language model and using adversarial training with a multi-lingual model, achieving first place in the SemEval-2020 Task 9 Hindi-English competition.
Code switching is a linguistic phenomenon that may occur within a multilingual setting where speakers share more than one language. With the increasing communication between groups with different languages, this phenomenon is more and more popular. However, there are little research and data in this area, especially in code-mixing sentiment classification. In this work, the domain transfer learning from state-of-the-art uni-language model ERNIE is tested on the code-mixing dataset, and surprisingly, a strong baseline is achieved. Furthermore, the adversarial training with a multi-lingual model is used to achieve 1st place of SemEval-2020 Task 9 Hindi-English sentiment classification competition.