AxBERT: An Interpretable Chinese Spelling Correction Method Driven by Associative Knowledge Network
This addresses the need for interpretable models in text correction tasks, particularly for Chinese language applications, though it is incremental as it builds on existing BERT methods.
The authors tackled the problem of uninterpretability in deep learning models for Chinese spelling correction by proposing AxBERT, which aligns BERT with an associative knowledge network to provide interpretable feature explanations, achieving improved precision on SIGHAN datasets.
Deep learning has shown promising performance on various machine learning tasks. Nevertheless, the uninterpretability of deep learning models severely restricts the usage domains that require feature explanations, such as text correction. Therefore, a novel interpretable deep learning model (named AxBERT) is proposed for Chinese spelling correction by aligning with an associative knowledge network (AKN). Wherein AKN is constructed based on the co-occurrence relations among Chinese characters, which denotes the interpretable statistic logic contrasted with uninterpretable BERT logic. And a translator matrix between BERT and AKN is introduced for the alignment and regulation of the attention component in BERT. In addition, a weight regulator is designed to adjust the attention distributions in BERT to appropriately model the sentence semantics. Experimental results on SIGHAN datasets demonstrate that AxBERT can achieve extraordinary performance, especially upon model precision compared to baselines. Our interpretable analysis, together with qualitative reasoning, can effectively illustrate the interpretability of AxBERT.