Learner Corpus: Collection and Application of the English Part of Incheon National University Multi-Language Learner Corpus (INU-MULC)

Abstract

This article describes the design, collection process, and applied research of the English part of Incheon National University Multi-language Learner Corpus, developed at a Korean university in a project that took place between 2018-2021. The total size of the English corpus was 415,000 words. The corpus consisted of recordings and transcripts of spoken corpus (including conversation and monologue) and written corpus. The spoken corpus had 116 conversations and 244 monologues of 345 students. The written corpus included 158 compositions. Although a number of foreign language learner corpora have been developed in the corpus field, spoken corpora, especially casual conversation data, are relatively rare considering their importance in the research of language acquisition. Unlike other corpora that mainly collect monologues, this corpus includes general conversation, which can be employed in various research of learners’ language including casual pronunciation, grammar, vocabulary selection, discourse markers, and conversation strategies. Findings from the research on learner evaluation and pronunciation suggest that researchers who are looking for large sized empirical data might benefit from this corpus in their research. This article will contribute to the development of various linguistic fields including language acquisition and newly building corpora.