Developing an Automated Analysis System of Lexical Richness for Chinese As a Second Language Text

Abstract

Abstract Lexical richness is an essential construct in second language writing research. The existing method to measure lexical richness in Chinese as a second language (CSL) research is inaccurate and time-consuming, which reduces the reliability and comparability of research conclusions and limits sample sizes. The present study develops a lexical richness automated analysis system for CSL text (i.e. the LRAAS) that can process CSL text documents in batches and produce fifteen main indices of lexical richness. The LRAAS has shown high reliability and accuracy in calculating the lexical richness of CSL text by comparing artificial and automated calculation results with different Chinese proficiency levels. Furthermore, the LRAAS will achieve greater accuracy as the lexicon and dynamic corpus training data expand. At present, the LRAAS is a reliable tool for CSL researchers analyzing learners’ lexical richness and has several potential applications in the field.