Computational Modelling of Plurality and Definiteness in Chinese Noun Phrases
This addresses a linguistic analysis problem for computational linguists and AI researchers, but it is incremental as it applies existing methods to a specific language dataset.
The paper tackled the problem of predicting omitted plurality and definiteness markers in Chinese noun phrases using context, and found that Chinese speakers frequently drop these markers, with computational models achieving specific performance metrics as reported.
Theoretical linguists have suggested that some languages (e.g., Chinese and Japanese) are "cooler" than other languages based on the observation that the intended meaning of phrases in these languages depends more on their contexts. As a result, many expressions in these languages are shortened, and their meaning is inferred from the context. In this paper, we focus on the omission of the plurality and definiteness markers in Chinese noun phrases (NPs) to investigate the predictability of their intended meaning given the contexts. To this end, we built a corpus of Chinese NPs, each of which is accompanied by its corresponding context, and by labels indicating its singularity/plurality and definiteness/indefiniteness. We carried out corpus assessments and analyses. The results suggest that Chinese speakers indeed drop plurality and definiteness markers very frequently. Building on the corpus, we train a bank of computational models using both classic machine learning models and state-of-the-art pre-trained language models to predict the plurality and definiteness of each NP. We report on the performance of these models and analyse their behaviours.