Multiple Admissibility in Language Learning: : Judging Grammaticality using Unlabeled Data

Show full item record



Permalink

http://hdl.handle.net/10138/317348

Citation

Katinskaia , A , Ivanova , S & Yangarber , R 2019 , Multiple Admissibility in Language Learning: Judging Grammaticality using Unlabeled Data . in T Erjavec , M Marcińczuk , P Nakov , J Piskorski , L Pivovarova , J Šnajder , J Steinberger & R Yangarber (eds) , The 7th Workshop on Balto-Slavic Natural Language Processing : Proceedings of the Workshop . The Association for Computational Linguistics , Stroudsburg , pp. 12-22 , Workshop on Balto-Slavic Natural Language Processing , Florence , Italy , 02/08/2019 . https://doi.org/10.18653/v1/W19-3702

Title: Multiple Admissibility in Language Learning: : Judging Grammaticality using Unlabeled Data
Author: Katinskaia, Anisia; Ivanova, Sardana; Yangarber, Roman
Editor: Erjavec, Tomaž; Marcińczuk, Michał; Nakov, Preslav; Piskorski, Jakub; Pivovarova, Lidia; Šnajder, Jan; Steinberger, Josef; Yangarber, Roman
Contributor: University of Helsinki, Department of Languages
University of Helsinki, Department of Computer Science
University of Helsinki, Department of Digital Humanities
Publisher: The Association for Computational Linguistics
Date: 2019-08
Language: eng
Number of pages: 11
Belongs to series: The 7th Workshop on Balto-Slavic Natural Language Processing Proceedings of the Workshop
ISBN: 978-1-950737-41-3
URI: http://hdl.handle.net/10138/317348
Abstract: We present our work on the problem of detection Multiple Admissibility (MA) in language learning. Multiple Admissibility occurs when more than one grammatical form of a word fits syntactically and semantically in a given context. In second-language education—in particular, in intelligent tutoring systems/computer-aided language learning (ITS/CALL), systems generate exercises automatically. MA implies that multiple alternative answers are possible. We treat the problem as a grammaticality judgement task. We train a neural network with an objective to label sentences as grammatical or ungrammatical, using a "simulated learner corpus": a dataset with correct text and with artificial errors, generated automatically. While MA occurs commonly in many languages, this paper focuses on learning Russian. We present a detailed classification of the types of constructions in Russian, in which MA is possible, and evaluate the model using a test set built from answers provided by users of the Revita language learning system.
Subject: 113 Computer and information sciences
Rights:


Files in this item

Total number of downloads: Loading...

Files Size Format View
W19_3702.pdf 256.7Kb PDF View/Open

This item appears in the following Collection(s)

Show full item record