POS Taggers’ Performance in Learner and Native Texts: Using the Korean Children’s English Learner Corpus
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This study evaluates the performances of different taggers—spaCy open–source, Stanford part–of–speech (POS), and CLAWS7—on two datasets: one with 88 essays from the Korean Children’s English Learner Corpus (L2 learners) and the other with 25 essays from native children (LUCY corpus). The total word count analyzed is 4,310 for L2 and 4,235 for the native dataset. POS tags assigned by each tagger were assessed using categories from previous studies, revealing spelling and learner errors. CLAWS7 demonstrated the highest accuracy rates of over 97.08% and 99.03% for L2 and native datasets, respectively, followed by Stanford POS (95.43% and 98.96%) and spaCy (83.60% and 87.27%). The study indicates that CLAWS7 excelled in accurately tagging text from the K–CHELC dataset, showcasing its effectiveness.
Abstract
This study evaluates the performances of different taggers—spaCy open–source, Stanford part–of–speech (POS), and CLAWS7—on two datasets: one with 88 essays from the Korean Children’s English Learner Corpus (L2 learners) and the other with 25 essays from native children (LUCY corpus). The total word count analyzed is 4,310 for L2 and 4,235 for the native dataset. POS tags assigned by each tagger were assessed using categories from previous studies, revealing spelling and learner errors. CLAWS7 demonstrated the highest accuracy rates of over 97.08% and 99.03% for L2 and native datasets, respectively, followed by Stanford POS (95.43% and 98.96%) and spaCy (83.60% and 87.27%). The study indicates that CLAWS7 excelled in accurately tagging text from the K–CHELC dataset, showcasing its effectiveness.
참고문헌 (17)
- Aberdour, M., "Achieving quality in open–source software", IEEE software, 24(1), pp.58~64 (2007)
- Aziz, R. A., "Tagging L2 Writing: Learner Errors and the Performance of an Automated Part–of–Speech Tagger", GEMA Online Journal of Language Studies, 19(3) (2019)
- de Haan, P., "Corpus linguistics and linguistic theory", pp.69~79 (2000), Brill
- Dıaz–Negrillo, A., "Towards interlanguage POS annotation for effective learner corpora in SLA and FLT", Language Forum, 36(1–2), pp.139~154 (2010)
- Grant, L, "Using computer–tagged linguistic features to describe L2 writing differences", Journal of second language writing, 9(2), pp.123~145 (2000)
- 김희재, "The Usage of Lexical Bundles in composition writings: The case of Korean Children Learners of English as a Foreign Language", 언어사실과 관점, 46, pp.109~132 (2019), 언어정보연구원[DOI]
- 김한샘, "한국어의 명사・부사 품사 통용 - 품사 통용어의 판별과 기술을 중심으로", 언어, 40(4), pp.601~618 (2015), 한국언어학회[DOI]
- Kohler, C. T., "African American English dialect and performance on nonword spelling and phonemic awareness tasks", American Journal of Speech–Language Pathology, 16, pp.157~168 (2007)
- Manning, C. D., "The Stanford CoreNLP natural language processing toolkit", Proceedings of 52nd annual meeting of the association for computational linguistics: system demonstrations, pp.55~60 (2014)
- Rayson, Paul, "The claws web tagger", ICAME Journal, 22, pp.121~123 (1998)
- Taylor, A., "The Penn Treebank: an overview, Treebanks: Building and using parsed corpora", 5~22 (2003)
- "The LUCY Corpus: Documentation" (2003)[URL]
- Spring, R., "The possibility of improving automated calculation of measures of lexical richness for EFL writing: A comparison of the LCA, NLTK and SpaCy tools", System, 106, pp.102770~ (2022)
- Toutanova, K, "Enriching the knowledge sources used in a maximum entropy part–of–speech tagger", 2000 Joint SIGDAT conference on Empirical methods in natural language processing and very large corpora, pp.63~70 (2000)
- Van Halteren, H., "Performance of taggers, Syntactic wordclass tagging", 81~94 (1999)
- Van Rooy, B., "The effect of learner errors on POS tag errors during automatic POS tagging", Southern African Linguistics and Applied Language Studies, 20(4), pp.325~335 (2002)
- Van Rooy, B., "An evaluation of three POS taggers for the tagging of the Tswana Learner English Corpus", Proceedings of the Corpus Linguistics 2003 conference, 16, pp.835~844 (2003)