Show simple item record

dc.contributor.authorBal, Mustafa
dc.date.accessioned2020-08-28T09:42:27Z
dc.date.created2019-05
dc.date.issued2019-08-23
dc.date.submitted2019
dc.identifier.citationBal, Mustafa. 2019. Predicting Non-Restrictive Noun Phrase Modifications Through Deep Semantic Analysis. Bachelor's thesis, Harvard College.
dc.identifier.urihttps://nrs.harvard.edu/URN-3:HUL.INSTREPOS:37364649*
dc.description.abstractThe difference between restrictive and non-restrictive modifier clauses has been well documented in linguistics. A model that can distinguish non-restrictive modifiers from restrictive modifiers can provide shorter sentences for Natural Language Processing applications3 and improve personal voice assistants in sounding more natural to users. Previous research has provided an annotated corpus and a relatively successful model that predicts non-restrictive modifiers in given sentences. However, this model suffers when faced with prepositional and adjectival modifiers. By utilizing this annotated corpus and learning from previously existing literature, we have made a model that can successfully predict non-restrictive noun phrase modifications through deeper semantic analysis while also performing better with prepositional and adjectival modifiers.
dc.description.sponsorshipComputer Science
dc.description.sponsorshipComputer Science
dc.format.mimetypeapplication/pdf
dc.language.isoen
dash.licenseLAA
dc.titlePredicting Non-Restrictive Noun Phrase Modifications Through Deep Semantic Analysis
dc.typeThesis or Dissertation
dash.depositing.authorBal, Mustafa
dc.date.available2020-08-28T09:42:27Z
thesis.degree.date2019
thesis.degree.grantorHarvard College
thesis.degree.grantorHarvard College
thesis.degree.levelUndergraduate
thesis.degree.levelUndergraduate
thesis.degree.nameAB
thesis.degree.nameAB
dc.type.materialtext
thesis.degree.departmentComputer Science
thesis.degree.departmentComputer Science
dash.identifier.vireo
dc.identifier.orcid0000-0003-4353-4873
dash.author.emailbalmustafa117@gmail.com


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record