Automatic Classification of the “De” Word Usage for Chinese as a Foreign Language

This paper proposed a word usage classification for “De” in Chinese as a secondary language by rule induction algorithm. Learning of Chinese characters and tone adaption are both essential and hard tasks for non-native speakers. The frequent terms, defined in morphosyntatic particle “De” with three characters {的, 得, 地}, is hard to learn for foreign learners due to the similar pronunciation and meaning. This investment illustrates a data-driven algorithm to classify the usages about the morphosyntatic particle “De” in Chinese learning. Rule induction is one of the most important techniques to learn the knowledge from data. Since regularities hidden in data are frequently expressed in terms of rules, rule induction is one of the fundamental tools for natural language processing and obtains a significant improvement in character selection. By the automatic rule induction process, 32 rules are adopted here to classify the character usage in morphosyntatic particle “De.” According to the experimental results, we find the proposed method can provide good enough performance to classify the character usages for morphosyntatic particle “De.”