Data Augmentation Methods for Anaphoric Zero Pronouns
This work addresses a domain-specific problem for natural language processing in pro-drop languages, but it is incremental as it builds on existing methods.
The paper tackled the limited resources for studying anaphoric zero pronoun interpretation in pro-drop languages by using five data augmentation methods to generate and detect them automatically, resulting in improved performance for two Arabic systems that surpassed state-of-the-art results.
In pro-drop language like Arabic, Chinese, Italian, Japanese, Spanish, and many others, unrealized (null) arguments in certain syntactic positions can refer to a previously introduced entity, and are thus called anaphoric zero pronouns. The existing resources for studying anaphoric zero pronoun interpretation are however still limited. In this paper, we use five data augmentation methods to generate and detect anaphoric zero pronouns automatically. We use the augmented data as additional training materials for two anaphoric zero pronoun systems for Arabic. Our experimental results show that data augmentation improves the performance of the two systems, surpassing the state-of-the-art results.