Paraphrasing Academic Text: A Study of Back-Translating Anatomy and Physiology with Transformers [Konferenzbericht] Paper presented at the International Conference on Artificial Intelligence in Education (22nd, 2021).

Autor/in	Olney, Andrew M.
Titel	Paraphrasing Academic Text: A Study of Back-Translating Anatomy and Physiology with Transformers [Konferenzbericht] Paper presented at the International Conference on Artificial Intelligence in Education (22nd, 2021).
Quelle	(2021), (6 Seiten) PDF als Volltext kostenfreie Datei Verfügbarkeit
Zusatzinformation	ORCID (Olney, Andrew M.) Weitere Informationen
Sprache	englisch
Dokumenttyp	gedruckt; online; Monographie
Schlagwörter	Translation; Language Processing; Error Analysis (Language); Grammar; Writing Skills; Models; Anatomy; Physiology; Textbooks; Science Instruction; Language Fluency; Computational Linguistics; Second Languages; Language Usage; Computer Software; Slavic Languages; Russian; English (Second Language); Welsh; Turkish; Indo European Languages; Chinese; Semitic Languages; Contrastive Linguistics; Written Language + Suchen Sie Ihr Suchwort? Sprachverarbeitung; Error analysis; Language; Fehleranalyse; Grammatik; Writing skill; Schreibfertigkeit; Analogiemodell; Anatomie; Physiologie; Textbook; Text book; Schulbuch; Lehrbuch; Teaching of science; Science education; Natural sciences Lessons; Naturwissenschaftlicher Unterricht; Language skill; Language skills; Sprachkompetenz; Linguistics; Computerlinguistik; Second language; Zweitsprache; Sprachgebrauch; Slawische Sprache; Russisch; English as second language; English; Second Language; Englisch als Zweitsprache; Walisisch; Türkisch; Indoeuropäisch; China; Chinesen; Arabisch; Hebräisch; Kontrastive Linguistik; Geschriebene Sprache
Abstract	This paper explores a general approach to paraphrase generation using a pre-trained seq2seq model fine-tuned using a back-translated anatomy and physiology textbook. Human ratings indicate that the paraphrase model generally preserved meaning and grammaticality/fluency: 70% of meaning ratings were above 75, and 40% of paraphrases were considered more grammatical/fluent than the originals. An error analysis suggests potential avenues for future work. [This paper was published in: "Proceedings of the 22nd International Conference on Artificial Intelligence in Education," edited by I. Roll, D. McNamara, S. Sosnovsky, R. Luckin, and V. Dimitrova, Springer International Publishing, 2021, pp. 279-284.] (As Provided).
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2024/1/01