Multi-lingual mathematical word problem generation using long short term memory networks with enhanced input features

Simple item page
dc.contributor.authorLiyanage V
dc.contributor.authorRanathunga S
dc.coverage.spatialMarseille, France
dc.date.accessioned2025-08-06T02:07:36Z
dc.date.available2025-08-06T02:07:36Z
dc.date.finish-date2020-05-16
dc.date.issued2020-01-01
dc.date.start-date2020-05-11
dc.description.abstractA Mathematical Word Problem (MWP) differs from a general textual representation due to the fact that it is comprised of numerical quantities and units, in addition to text. Therefore, MWP generation should be carefully handled. When it comes to multi-lingual MWP generation, language specific morphological and syntactic features become additional constraints. Standard template-based MWP generation techniques are incapable of identifying these language specific constraints, particularly in morphologically rich yet low resource languages such as Sinhala and Tamil. This paper presents the use of a Long Short Term Memory (LSTM) network that is capable of generating elementary level MWPs, while satisfying the aforementioned constraints. Our approach feeds a combination of character embeddings, word embeddings, and Part of Speech (POS) tag embeddings to the LSTM, in which attention is provided for numerical values and units. We trained our model for three languages, English, Sinhala and Tamil using separate MWP datasets. Irrespective of the language and the type of the MWP, our model could generate accurate single sentenced and multi sentenced problems. Accuracy reported in terms of average BLEU score for English, Sinhala and Tamil languages were 22.97%, 24.49% and 20.74%, respectively.
dc.description.confidentialfalse
dc.format.pagination4709-4716
dc.identifier.citationLiyanage V, Ranathunga S. (2020). Multi-lingual mathematical word problem generation using long short term memory networks with enhanced input features. Lrec 2020 12th International Conference on Language Resources and Evaluation Conference Proceedings. (pp. 4709-4716). European Language Resources Association (ELRA).
dc.identifier.elements-typec-conference-paper-in-proceedings
dc.identifier.urihttps://mro.massey.ac.nz/handle/10179/73301
dc.publisherEuropean Language Resources Association (ELRA)
dc.publisher.urihttp://lrec-conf.org/proceedings/lrec2020/pdf/2020.lrec-1.579.pdf
dc.rights(c) The author/sen
dc.rights.licenseCC BY-NCen
dc.rights.urihttps://creativecommons.org/licenses/by-nc/4.0/deed.enen
dc.source.journalLrec 2020 12th International Conference on Language Resources and Evaluation Conference Proceedings
dc.source.name-of-conference12th Conference on Language Resources and Evaluation (LREC 2020)
dc.subjectEmbeddings
dc.subjectLanguage Generation
dc.subjectLow- resource Languages
dc.subjectLSTM
dc.subjectMathematical Word Problem
dc.titleMulti-lingual mathematical word problem generation using long short term memory networks with enhanced input features
dc.typeconference
pubs.elements-id488655
pubs.organisational-groupOther
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
488655 PDF.pdf
Size:
407.73 KB
Format:
Adobe Portable Document Format
Description:
Published version.pdf
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
9.22 KB
Format:
Plain Text
Description:
Download
Collections
Conference Papers