Japanese Lexical Complexity for Non-Native Readers: A New Dataset
This work addresses the problem of assisting non-native readers in Japanese by providing a new dataset, though it is incremental as it extends existing LCP methods to a new language.
The authors tackled the lack of lexical complexity prediction resources for Japanese by constructing the first Japanese LCP dataset with L1-specific scores, and they demonstrated a BERT-based system's effectiveness in baseline experiments.
Lexical complexity prediction (LCP) is the task of predicting the complexity of words in a text on a continuous scale. It plays a vital role in simplifying or annotating complex words to assist readers. To study lexical complexity in Japanese, we construct the first Japanese LCP dataset. Our dataset provides separate complexity scores for Chinese/Korean annotators and others to address the readers' L1-specific needs. In the baseline experiment, we demonstrate the effectiveness of a BERT-based system for Japanese LCP.