Incorporating Chinese Characters of Words for Lexical Sememe Prediction
This work improves sememe annotation efficiency and consistency for Chinese NLP tasks, but it is incremental as it builds on existing methods by adding character information.
The paper tackles the lexical sememe prediction task for Chinese by addressing the issue of low-frequency and out-of-vocabulary words, proposing a framework that combines internal character and external context information, which outperforms state-of-the-art baselines by a large margin and maintains robust performance for low-frequency words.
Sememes are minimum semantic units of concepts in human languages, such that each word sense is composed of one or multiple sememes. Words are usually manually annotated with their sememes by linguists, and form linguistic common-sense knowledge bases widely used in various NLP tasks. Recently, the lexical sememe prediction task has been introduced. It consists of automatically recommending sememes for words, which is expected to improve annotation efficiency and consistency. However, existing methods of lexical sememe prediction typically rely on the external context of words to represent the meaning, which usually fails to deal with low-frequency and out-of-vocabulary words. To address this issue for Chinese, we propose a novel framework to take advantage of both internal character information and external context information of words. We experiment on HowNet, a Chinese sememe knowledge base, and demonstrate that our framework outperforms state-of-the-art baselines by a large margin, and maintains a robust performance even for low-frequency words.