End to End Chinese Lexical Fusion Recognition with Sememe Knowledge
This work addresses a specific problem in natural language processing for Chinese language understanding, but it is incremental as it adapts existing methods to a new task.
The paper tackles Chinese lexical fusion recognition, a new task similar to coreference recognition, by proposing an end-to-end joint model that uses BERT and sememe knowledge from HowNet, and demonstrates its effectiveness on a manually annotated benchmark dataset.
In this paper, we present Chinese lexical fusion recognition, a new task which could be regarded as one kind of coreference recognition. First, we introduce the task in detail, showing the relationship with coreference recognition and differences from the existing tasks. Second, we propose an end-to-end joint model for the task, which exploits the state-of-the-art BERT representations as encoder, and is further enhanced with the sememe knowledge from HowNet by graph attention networks. We manually annotate a benchmark dataset for the task and then conduct experiments on it. Results demonstrate that our joint model is effective and competitive for the task. Detailed analysis is offered for comprehensively understanding the new task and our proposed model.