{"id":{"repo_id":"reykjavik","oai_identifier":"oai:skemman.is:1946/39413"},"canonical_url":"https://search.dev.ndltd.org/etd/reykjavik/oai:skemman.is:1946/39413","repository":{"repo_id":"reykjavik","name":"Reykjavík University","base_url":"https://skemman.is/oai/request"},"display":{"title":"Adapting and improving a shallow parser","abstract":"IceNLP is a set of tools for processing and analysing the Icelandic language. One of these tools is IceParser, a shallow parser, whose purpose is to annotate flat phrase structure and syntactic functions. The input to IceParser are tokenised sentences, tagged with part-of-speech (POS) tags from the Icelandic POS tagset. This tagset was revised recently which means that many tools need to be revised to accurately handle the changes. In this project, we, first, present the changes made to IceParser with regard to the tagset revision. Second, we describe improvements made to individual components of IceParser, for the purpose of increasing its precision and coverage. Third, we perform an evaluation of the functionality of IceParser after the changes and, finally, perform preliminary error analysis.","abstract_html":"IceNLP is a set of tools for processing and analysing the Icelandic language. One of these tools is IceParser, a shallow parser, whose purpose is to annotate flat phrase structure and syntactic functions. The input to IceParser are tokenised sentences, tagged with part-of-speech (POS) tags from the Icelandic POS tagset. This tagset was revised recently which means that many tools need to be revised to accurately handle the changes. In this project, we, first, present the changes made to IceParser with regard to the tagset revision. Second, we describe improvements made to individual components of IceParser, for the purpose of increasing its precision and coverage. Third, we perform an evaluation of the functionality of IceParser after the changes and, finally, perform preliminary error analysis.","abstract_has_math":false,"creators":["Gunnar Pálsson 1993-"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Háskólinn í Reykjavík"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2021,"date_issued":"2021-06-22T15:42:08Z","date_published":"2021-06-22T15:42:08Z","updated_at":"2026-07-27T20:41:33Z","subjects":["Tölvunarfræði","Meistaraprófsritgerðir","Hugbúnaðargerð","Máltækni","Textagreining","Íslenska","Computer science","Software development","Natural language processing (Computer science)","Text analysis","Icelandic"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/1946/39413","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Háskólinn í Reykjavík"]},{"key":"dc:creator","label":"Author","values":["Gunnar Pálsson 1993-"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2021-06-22T15:42:04Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2021-06-22T15:42:04Z"]},{"key":"dc:date.issued","label":"Date","values":["2021-06-22T15:42:08Z"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Tölvunarfræði","Meistaraprófsritgerðir","Hugbúnaðargerð","Máltækni","Textagreining","Íslenska","Computer science","Software development","Natural language processing (Computer science)","Text analysis","Icelandic"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://hdl.handle.net/1946/39413"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["IceNLP is a set of tools for processing and analysing the Icelandic language. One of these tools is IceParser, a shallow parser, whose purpose is to annotate flat phrase structure and syntactic functions. The input to IceParser are tokenised sentences, tagged with part-of-speech (POS) tags from the Icelandic POS tagset. This tagset was revised recently which means that many tools need to be revised to accurately handle the changes. In this project, we, first, present the changes made to IceParser with regard to the tagset revision. Second, we describe improvements made to individual components of IceParser, for the purpose of increasing its precision and coverage. Third, we perform an evaluation of the functionality of IceParser after the changes and, finally, perform preliminary error analysis.","IceNLP er safn hugbúnaðartóla til að vinna með og greina íslenskan texta. Eitt þessara tóla er IceParser, hlutaþáttari með þann tilgang að greina setningarliði og setningafræðileg hlutverk. Inntak IceParser eru tilreiddar setningar sem hafa verið markaðar samkvæmt íslenska markamenginu. Íslenska markamengið var nýlega uppfært og þurfa tól IceNLP uppfærslu til að taka á breytingum markamengisins vegna þess. Í þessari ritgerð kynnum við breytingar sem hafa verið gerðar á IceParser. Við kynnum einnig almennar umbætur á einstökum hlutum IceParser sem eru gerðar í þeim tilgangi að bæta nákvæmni og griphlutfall. Við framkvæmum mælingu á virkni IceParser eftir breytingarnar og að lokum framkvæmum við villugreiningu."]},{"key":"dc:title","label":"Title","values":["Adapting and improving a shallow parser","Aðlögun og umbætur á hlutaþáttara"]}]}],"canonical_facts":{"dc:contributor":["Háskólinn í Reykjavík"],"dc:creator":["Gunnar Pálsson 1993-"],"dc:date.accessioned":["2021-06-22T15:42:04Z"],"dc:date.available":["2021-06-22T15:42:04Z"],"dc:date.issued":["2021-06-22T15:42:08Z"],"dc:description.abstract":["IceNLP is a set of tools for processing and analysing the Icelandic language. One of these tools is IceParser, a shallow parser, whose purpose is to annotate flat phrase structure and syntactic functions. The input to IceParser are tokenised sentences, tagged with part-of-speech (POS) tags from the Icelandic POS tagset. This tagset was revised recently which means that many tools need to be revised to accurately handle the changes. In this project, we, first, present the changes made to IceParser with regard to the tagset revision. Second, we describe improvements made to individual components of IceParser, for the purpose of increasing its precision and coverage. Third, we perform an evaluation of the functionality of IceParser after the changes and, finally, perform preliminary error analysis.","IceNLP er safn hugbúnaðartóla til að vinna með og greina íslenskan texta. Eitt þessara tóla er IceParser, hlutaþáttari með þann tilgang að greina setningarliði og setningafræðileg hlutverk. Inntak IceParser eru tilreiddar setningar sem hafa verið markaðar samkvæmt íslenska markamenginu. Íslenska markamengið var nýlega uppfært og þurfa tól IceNLP uppfærslu til að taka á breytingum markamengisins vegna þess. Í þessari ritgerð kynnum við breytingar sem hafa verið gerðar á IceParser. Við kynnum einnig almennar umbætur á einstökum hlutum IceParser sem eru gerðar í þeim tilgangi að bæta nákvæmni og griphlutfall. Við framkvæmum mælingu á virkni IceParser eftir breytingarnar og að lokum framkvæmum við villugreiningu."],"dc:identifier.uri":["http://hdl.handle.net/1946/39413"],"dc:language.iso":["en"],"dc:subject":["Tölvunarfræði","Meistaraprófsritgerðir","Hugbúnaðargerð","Máltækni","Textagreining","Íslenska","Computer science","Software development","Natural language processing (Computer science)","Text analysis","Icelandic"],"dc:title":["Adapting and improving a shallow parser","Aðlögun og umbætur á hlutaþáttara"],"dc:type":["Thesis"]},"updated_at":"2026-07-27T20:41:33Z"}