MCSCSet: A Specialist-annotated Dataset for Medical-domain Chinese Spelling Correction

TL;DR

This paper introduces MCSCSet, a large, specialist-annotated dataset for Chinese medical spelling correction, highlighting the unique challenges in medical contexts and benchmarking models to improve accuracy.

Contribution

The paper presents MCSCSet, a novel large-scale medical-domain Chinese spelling correction dataset with expert annotations and a medical confusion set, addressing domain-specific challenges.

Findings

Significant performance gaps between open-domain and medical-domain CSC models.

MCSCSet enables automatic generation of medical misspelling datasets.

Benchmarking results establish baselines for future research.

Abstract

Chinese Spelling Correction (CSC) is gaining increasing attention due to its promise of automatically detecting and correcting spelling errors in Chinese texts. Despite its extensive use in many applications, like search engines and optical character recognition systems, little has been explored in medical scenarios in which complex and uncommon medical entities are easily misspelled. Correcting the misspellings of medical entities is arguably more difficult than those in the open domain due to its requirements of specificdomain knowledge. In this work, we define the task of Medical-domain Chinese Spelling Correction and propose MCSCSet, a large scale specialist-annotated dataset that contains about 200k samples. In contrast to the existing open-domain CSC datasets, MCSCSet involves: i) extensive real-world medical queries collected from Tencent Yidian, ii) corresponding misspelled sentences manually annotated by medical specialists. To ensure automated dataset curation, MCSCSet further offers a medical confusion set consisting of the commonly misspelled characters of given Chinese medical terms. This enables one to create the medical misspelling dataset automatically. Extensive empirical studies have shown significant performance gaps between the open-domain and medical-domain spelling correction, highlighting the need to develop high-quality datasets that allow for Chinese spelling correction in specific domains. Moreover, our work benchmarks several representative Chinese spelling correction models, establishing baselines for future work.