{"id":{"repo_id":"cornell","oai_identifier":"oai:ecommons.cornell.edu:1813/47892"},"canonical_url":"https://search.ndltd.org/etd/cornell/oai:ecommons.cornell.edu:1813/47892","repository":{"repo_id":"cornell","name":"Cornell University","base_url":"https://ecommons.cornell.edu/server/oai/request"},"display":{"title":"Deep Sequential and Structural Neural Models of Compositionality","abstract":"Recent advances in deep learning have provided fruitful applications for natural language processing (NLP) tasks. One key advance was the invention of word vectors, representing every word in a dense, low-dimensional vector space. Even though word vectors provide very strong results for word level NLP tasks, producing appropriate representation for phrases and sentences is still an open research problem. In this dissertation, we focus on compositional approaches to representation learning. In particular, we employ the notions of compositionality in which the sequence or structure information is utilized, via recurrent or recursive neural networks. We investigate the effectiveness of such approaches for specific natural language understanding tasks including opinion mining and sentiment analysis, and extend some of the approaches to provide better representation hierarchies. In particular, we propose two novel variants: bidirectional recursive neural networks, which are capable of producing context-dependent structural representations and deep recursive neural networks, which provide representation hierarchies in the structural setting. Additionally, we qualitatively investigate such models, and describe how they relate to alternative compositional approaches. Finally, we discuss challenges in interpretation and understanding of compositional neural models, propose simple tools for visualization, and perform exploratory analyses over features learned by such a model.","abstract_html":"Recent advances in deep learning have provided fruitful applications for natural language processing (NLP) tasks. One key advance was the invention of word vectors, representing every word in a dense, low-dimensional vector space. Even though word vectors provide very strong results for word level NLP tasks, producing appropriate representation for phrases and sentences is still an open research problem. In this dissertation, we focus on compositional approaches to representation learning. In particular, we employ the notions of compositionality in which the sequence or structure information is utilized, via recurrent or recursive neural networks. We investigate the effectiveness of such approaches for specific natural language understanding tasks including opinion mining and sentiment analysis, and extend some of the approaches to provide better representation hierarchies. In particular, we propose two novel variants: bidirectional recursive neural networks, which are capable of producing context-dependent structural representations and deep recursive neural networks, which provide representation hierarchies in the structural setting. Additionally, we qualitatively investigate such models, and describe how they relate to alternative compositional approaches. Finally, we discuss challenges in interpretation and understanding of compositional neural models, propose simple tools for visualization, and perform exploratory analyses over features learned by such a model.","abstract_has_math":false,"creators":["Irsoy, Ozan"],"institution":"Cornell University","degree_name":"Ph. D., Computer Science","degree_level":"Doctor of Philosophy","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":["Cardie, Claire T"],"committee_members":["Woodard, Dawn B.","Kleinberg, Robert David"],"year":2017,"date_issued":"2017-01-30","date_published":"2017-01-30","updated_at":"2026-11-19T03:19:15Z","subjects":["neural networks","Computer science","machine learning","natural language processing","Deep Learning"],"languages":["en_US"],"rights":["Attribution 4.0 International"],"rights_urls":["https://creativecommons.org/licenses/by/4.0/"],"identifier_entries":[{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.7298/X46971JG"],"render_values":[{"text":"https://doi.org/10.7298/X46971JG","href":"https://doi.org/10.7298/X46971JG","code":true}]},{"key":"dc:identifier.bibid","label":"Dc Identifier Bibid","values":["9906139"],"render_values":[{"text":"9906139","href":null,"code":true}]},{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["ProQuest Submission ID: 10131","ProQuest Publication ID: 10253515"],"render_values":[{"text":"ProQuest Submission ID: 10131","href":null,"code":true},{"text":"ProQuest Publication ID: 10253515","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1813/47892","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"https://hdl.handle.net/1813/47892","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://ecommons.cornell.edu/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Aecommons.cornell.edu%3A1813%2F47892","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.chair","label":"Chair","values":["Cardie, Claire T"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Woodard, Dawn B.","Kleinberg, Robert David"]},{"key":"dc:creator","label":"Author","values":["Irsoy, Ozan"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2017-04-04T20:28:31Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2017-04-04T20:28:31Z"]},{"key":"dc:date.issued","label":"Date","values":["2017-01-30"]},{"key":"dc:type","label":"Dc Type","values":["dissertation or thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Doctor of Philosophy"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph. D., Computer Science"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Cornell University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["neural networks","Computer science","machine learning","natural language processing","Deep Learning"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en_US"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution 4.0 International"]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://creativecommons.org/licenses/by/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.bibid","label":"Dc Identifier Bibid","values":["9906139"]},{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.7298/X46971JG"]},{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["ProQuest Submission ID: 10131","ProQuest Publication ID: 10253515"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1813/47892"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"cerif:openaire","label":"Cerif Openaire","values":["<Publication xmlns=\"https://www.openaire.eu/cerif-profile/1.1/\" id=\"c67a70e2-a5a9-42aa-bee0-84d5d4a31872\"> <Type xmlns=\"https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types\">http://purl.org/coar/resource_type/c_1843</Type> <Language>en_US</Language> <Title>Deep Sequential and Structural Neural Models of Compositionality</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2017-01-30</PublicationDate> <DOI>https://doi.org/10.7298/X46971JG</DOI> <Authors> <Author> <DisplayName>Irsoy, Ozan</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <OrgUnit /> </Publisher> </Publishers> <License>https://creativecommons.org/licenses/by/4.0/</License> <Keyword>neural networks</Keyword> <Keyword>Computer science</Keyword> <Keyword>machine learning</Keyword> <Keyword>natural language processing</Keyword> <Keyword>Deep Learning</Keyword> <Abstract>Recent advances in deep learning have provided fruitful applications for natural language processing (NLP) tasks. One key advance was the invention of word vectors, representing every word in a dense, low-dimensional vector space. Even though word vectors provide very strong results for word level NLP tasks, producing appropriate representation for phrases and sentences is still an open research problem. In this dissertation, we focus on compositional approaches to representation learning. In particular, we employ the notions of compositionality in which the sequence or structure information is utilized, via recurrent or recursive neural networks. We investigate the effectiveness of such approaches for specific natural language understanding tasks including opinion mining and sentiment analysis, and extend some of the approaches to provide better representation hierarchies. In particular, we propose two novel variants: bidirectional recursive neural networks, which are capable of producing context-dependent structural representations and deep recursive neural networks, which provide representation hierarchies in the structural setting. Additionally, we qualitatively investigate such models, and describe how they relate to alternative compositional approaches. Finally, we discuss challenges in interpretation and understanding of compositional neural models, propose simple tools for visualization, and perform exploratory analyses over features learned by such a model.</Abstract> <Access xmlns=\"http://purl.org/coar/access_right\" > </Access> </Publication>"]},{"key":"cris:virtual.author","label":"Cris Virtual Author","values":["Irsoy, Ozan"]},{"key":"cris:virtualsource.collection","label":"Cris Virtualsource Collection","values":["5893a6ea-7af3-41d7-abc6-04bcd26ab5df"]},{"key":"dc:description.abstract","label":"Abstract","values":["Recent advances in deep learning have provided fruitful applications for natural language processing (NLP) tasks. One key advance was the invention of word vectors, representing every word in a dense, low-dimensional vector space. Even though word vectors provide very strong results for word level NLP tasks, producing appropriate representation for phrases and sentences is still an open research problem. In this dissertation, we focus on compositional approaches to representation learning. In particular, we employ the notions of compositionality in which the sequence or structure information is utilized, via recurrent or recursive neural networks. We investigate the effectiveness of such approaches for specific natural language understanding tasks including opinion mining and sentiment analysis, and extend some of the approaches to provide better representation hierarchies. In particular, we propose two novel variants: bidirectional recursive neural networks, which are capable of producing context-dependent structural representations and deep recursive neural networks, which provide representation hierarchies in the structural setting. Additionally, we qualitatively investigate such models, and describe how they relate to alternative compositional approaches. Finally, we discuss challenges in interpretation and understanding of compositional neural models, propose simple tools for visualization, and perform exploratory analyses over features learned by such a model."]},{"key":"dc:description.collection","label":"Dc Description Collection","values":["Cornell Theses and Dissertations"]},{"key":"dc:format.mimetype","label":"Dc Format Mimetype","values":["application/pdf"]},{"key":"dcterms:license","label":"Dcterms License","values":["https://hdl.handle.net/1813/59810"]},{"key":"dc:title","label":"Title","values":["Deep Sequential and Structural Neural Models of Compositionality"]},{"key":"dspace:entity.type","label":"Dspace Entity Type","values":["Publication"]},{"key":"others:access_status","label":"Others Access Status","values":["open.access"]}]}],"canonical_facts":{"cerif:openaire":["<Publication xmlns=\"https://www.openaire.eu/cerif-profile/1.1/\" id=\"c67a70e2-a5a9-42aa-bee0-84d5d4a31872\"> <Type xmlns=\"https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types\">http://purl.org/coar/resource_type/c_1843</Type> <Language>en_US</Language> <Title>Deep Sequential and Structural Neural Models of Compositionality</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2017-01-30</PublicationDate> <DOI>https://doi.org/10.7298/X46971JG</DOI> <Authors> <Author> <DisplayName>Irsoy, Ozan</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <OrgUnit /> </Publisher> </Publishers> <License>https://creativecommons.org/licenses/by/4.0/</License> <Keyword>neural networks</Keyword> <Keyword>Computer science</Keyword> <Keyword>machine learning</Keyword> <Keyword>natural language processing</Keyword> <Keyword>Deep Learning</Keyword> <Abstract>Recent advances in deep learning have provided fruitful applications for natural language processing (NLP) tasks. One key advance was the invention of word vectors, representing every word in a dense, low-dimensional vector space. Even though word vectors provide very strong results for word level NLP tasks, producing appropriate representation for phrases and sentences is still an open research problem. In this dissertation, we focus on compositional approaches to representation learning. In particular, we employ the notions of compositionality in which the sequence or structure information is utilized, via recurrent or recursive neural networks. We investigate the effectiveness of such approaches for specific natural language understanding tasks including opinion mining and sentiment analysis, and extend some of the approaches to provide better representation hierarchies. In particular, we propose two novel variants: bidirectional recursive neural networks, which are capable of producing context-dependent structural representations and deep recursive neural networks, which provide representation hierarchies in the structural setting. Additionally, we qualitatively investigate such models, and describe how they relate to alternative compositional approaches. Finally, we discuss challenges in interpretation and understanding of compositional neural models, propose simple tools for visualization, and perform exploratory analyses over features learned by such a model.</Abstract> <Access xmlns=\"http://purl.org/coar/access_right\" > </Access> </Publication>"],"cris:virtual.author":["Irsoy, Ozan"],"cris:virtualsource.collection":["5893a6ea-7af3-41d7-abc6-04bcd26ab5df"],"dc:contributor.chair":["Cardie, Claire T"],"dc:contributor.committeemember":["Woodard, Dawn B.","Kleinberg, Robert David"],"dc:creator":["Irsoy, Ozan"],"dc:date.accessioned":["2017-04-04T20:28:31Z"],"dc:date.available":["2017-04-04T20:28:31Z"],"dc:date.issued":["2017-01-30"],"dc:description.abstract":["Recent advances in deep learning have provided fruitful applications for natural language processing (NLP) tasks. One key advance was the invention of word vectors, representing every word in a dense, low-dimensional vector space. Even though word vectors provide very strong results for word level NLP tasks, producing appropriate representation for phrases and sentences is still an open research problem. In this dissertation, we focus on compositional approaches to representation learning. In particular, we employ the notions of compositionality in which the sequence or structure information is utilized, via recurrent or recursive neural networks. We investigate the effectiveness of such approaches for specific natural language understanding tasks including opinion mining and sentiment analysis, and extend some of the approaches to provide better representation hierarchies. In particular, we propose two novel variants: bidirectional recursive neural networks, which are capable of producing context-dependent structural representations and deep recursive neural networks, which provide representation hierarchies in the structural setting. Additionally, we qualitatively investigate such models, and describe how they relate to alternative compositional approaches. Finally, we discuss challenges in interpretation and understanding of compositional neural models, propose simple tools for visualization, and perform exploratory analyses over features learned by such a model."],"dc:description.collection":["Cornell Theses and Dissertations"],"dc:format.mimetype":["application/pdf"],"dc:identifier.bibid":["9906139"],"dc:identifier.doi":["https://doi.org/10.7298/X46971JG"],"dc:identifier.other":["ProQuest Submission ID: 10131","ProQuest Publication ID: 10253515"],"dc:identifier.uri":["https://hdl.handle.net/1813/47892"],"dc:language.iso":["en_US"],"dc:rights":["Attribution 4.0 International"],"dc:rights.uri":["https://creativecommons.org/licenses/by/4.0/"],"dc:subject":["neural networks","Computer science","machine learning","natural language processing","Deep Learning"],"dcterms:license":["https://hdl.handle.net/1813/59810"],"dc:title":["Deep Sequential and Structural Neural Models of Compositionality"],"dc:type":["dissertation or thesis"],"dspace:entity.type":["Publication"],"others:access_status":["open.access"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["Doctor of Philosophy"],"thesis:degree_name":["Ph. D., Computer Science"],"thesis:institution_name":["Cornell University"]},"updated_at":"2026-11-19T03:19:15Z"}