Publicação
Semantic perspectives for learning over biomedical knowledge graphs
| Resumo: | Knowledge graphs represent an unparalleled opportunity for machine learning in the biomedical domain, given their ability to enrich data with meaningful context through semantic representations, such as knowledge graph embeddings and semantic similarity. However, the specificity of many biomedical tasks contrasts with the broad domains covered by large and successful biomedical knowledge graphs that describe entities according to several perspectives — semantic aspects. This is particularly challenging for predicting specific relations between entities described in the knowledge graph when the graph itself does not encode these relations. Current semantic representation methods consider the knowledge graph as a whole, ignoring the different semantic aspects. This thesis hypothesizes that semantic representations that are able to distinguish semantic aspects can improve the performance and explainability of biomedical relation prediction tasks. This work investigated different paradigms for defining semantic aspects based on classes and properties and developed multiple semantic representation techniques for both individual entities and entity pairs, with a focus on their explainability. Extensive experiments in proteinprotein interaction and gene-disease association predictions supported the empirical evaluation of the proposed methods and demonstrated that semantic aspect-oriented representations improve both predictive performance and explainability, fostering biomedical research. This work further highlights that in complex and multi-disciplinary domains, where a single knowledge graph is used to support a wide variety of tasks, it is essential to shift from viewing knowledge graphs as a whole to focusing on specific semantic perspectives. |
|---|---|
| Autores principais: | Sousa, Rita Isabel Torres de |
| Assunto: | Knowledge graph Semantic similarity Knowledge graph embedding Machine learning Biomedical application Grafo de conhecimento Semelhança semântica Embedding de grafos de conhecimento Aprendizagem automática Aplicação biomédica |
| Ano: | 2024 |
| País: | Portugal |
| Tipo de documento: | tese de doutoramento |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório da Universidade de Lisboa |
| _version_ | 1866808976267542528 |
|---|---|
| author | Sousa, Rita Isabel Torres de |
| author_facet | Sousa, Rita Isabel Torres de |
| author_role | author |
| contributor_name_str_mv | Pesquita, Cátia Luísa Santana Calisto Silva, Sara Guilherme Oliveira da Repositório Científico de Acesso Aberto da ULisboa |
| country_str | PT |
| creators_json_txt | [{\"Person.name\":\"Sousa, Rita Isabel Torres de\"}] |
| datacite.contributors.contributor.contributorName.fl_str_mv | Pesquita, Cátia Luísa Santana Calisto Silva, Sara Guilherme Oliveira da Repositório Científico de Acesso Aberto da ULisboa |
| datacite.creators.creator.creatorName.fl_str_mv | Sousa, Rita Isabel Torres de |
| datacite.date.Accepted.fl_str_mv | 2024-07-01T00:00:00Z |
| datacite.date.available.fl_str_mv | 2025-02-18T15:39:03Z |
| datacite.date.embargoed.fl_str_mv | 2025-02-18T15:39:03Z |
| datacite.rights.fl_str_mv | http://purl.org/coar/access_right/c_abf2 |
| datacite.subjects.subject.fl_str_mv | Knowledge graph Semantic similarity Knowledge graph embedding Machine learning Biomedical application Grafo de conhecimento Semelhança semântica Embedding de grafos de conhecimento Aprendizagem automática Aplicação biomédica |
| datacite.titles.title.fl_str_mv | Semantic perspectives for learning over biomedical knowledge graphs |
| dc.contributor.none.fl_str_mv | Pesquita, Cátia Luísa Santana Calisto Silva, Sara Guilherme Oliveira da Repositório Científico de Acesso Aberto da ULisboa |
| dc.creator.none.fl_str_mv | Sousa, Rita Isabel Torres de |
| dc.date.Accepted.fl_str_mv | 2024-07-01T00:00:00Z |
| dc.date.available.fl_str_mv | 2025-02-18T15:39:03Z |
| dc.date.embargoed.fl_str_mv | 2025-02-18T15:39:03Z |
| dc.format.none.fl_str_mv | application/pdf |
| dc.identifier.none.fl_str_mv | http://hdl.handle.net/10400.5/98530 |
| dc.language.none.fl_str_mv | eng |
| dc.rights.none.fl_str_mv | http://purl.org/coar/access_right/c_abf2 |
| dc.subject.none.fl_str_mv | Knowledge graph Semantic similarity Knowledge graph embedding Machine learning Biomedical application Grafo de conhecimento Semelhança semântica Embedding de grafos de conhecimento Aprendizagem automática Aplicação biomédica |
| dc.title.fl_str_mv | Semantic perspectives for learning over biomedical knowledge graphs |
| dc.type.none.fl_str_mv | http://purl.org/coar/resource_type/c_db06 |
| description | Knowledge graphs represent an unparalleled opportunity for machine learning in the biomedical domain, given their ability to enrich data with meaningful context through semantic representations, such as knowledge graph embeddings and semantic similarity. However, the specificity of many biomedical tasks contrasts with the broad domains covered by large and successful biomedical knowledge graphs that describe entities according to several perspectives — semantic aspects. This is particularly challenging for predicting specific relations between entities described in the knowledge graph when the graph itself does not encode these relations. Current semantic representation methods consider the knowledge graph as a whole, ignoring the different semantic aspects. This thesis hypothesizes that semantic representations that are able to distinguish semantic aspects can improve the performance and explainability of biomedical relation prediction tasks. This work investigated different paradigms for defining semantic aspects based on classes and properties and developed multiple semantic representation techniques for both individual entities and entity pairs, with a focus on their explainability. Extensive experiments in proteinprotein interaction and gene-disease association predictions supported the empirical evaluation of the proposed methods and demonstrated that semantic aspect-oriented representations improve both predictive performance and explainability, fostering biomedical research. This work further highlights that in complex and multi-disciplinary domains, where a single knowledge graph is used to support a wide variety of tasks, it is essential to shift from viewing knowledge graphs as a whole to focusing on specific semantic perspectives. |
| dirty | 0 |
| eu_rights_str_mv | openAccess |
| format | doctoralThesis |
| fulltext.url.fl_str_mv | https://repositorio.ulisboa.pt/bitstreams/9daa1ab8-c388-436e-ad8d-00927f30e953/download |
| funding.funder.alternateName_str_mv | FCT |
| funding.funder.identifier_str_mv | http://doi.org/10.13039/501100001871 |
| funding.funder.name_str_mv | Fundação para a Ciência e a Tecnologia |
| id | ul_f302a7992fca9ebbcb48ff14b282eb21 |
| identifier.url.fl_str_mv | http://hdl.handle.net/10400.5/98530 |
| instacron_str | ul |
| institution | Universidade de Lisboa |
| instname_str | Universidade de Lisboa |
| language | eng |
| network_acronym_str | ul |
| network_name_str | Repositório da Universidade de Lisboa |
| oai_identifier_str | oai:repositorio.ulisboa.pt:10400.5/98530 |
| organization_str_mv | urn:organizationAcronym:ul |
| person_str_mv | Sousa, Rita Isabel Torres de |
| publishDate | 2024 |
| reponame_str | Repositório da Universidade de Lisboa |
| repository_id_str | urn:repositoryAcronym:ul |
| service_str_mv | urn:repositoryAcronym:ul |
| spelling | engpt_PTKnowledge graphs represent an unparalleled opportunity for machine learning in the biomedical domain, given their ability to enrich data with meaningful context through semantic representations, such as knowledge graph embeddings and semantic similarity. However, the specificity of many biomedical tasks contrasts with the broad domains covered by large and successful biomedical knowledge graphs that describe entities according to several perspectives — semantic aspects. This is particularly challenging for predicting specific relations between entities described in the knowledge graph when the graph itself does not encode these relations. Current semantic representation methods consider the knowledge graph as a whole, ignoring the different semantic aspects. This thesis hypothesizes that semantic representations that are able to distinguish semantic aspects can improve the performance and explainability of biomedical relation prediction tasks. This work investigated different paradigms for defining semantic aspects based on classes and properties and developed multiple semantic representation techniques for both individual entities and entity pairs, with a focus on their explainability. Extensive experiments in proteinprotein interaction and gene-disease association predictions supported the empirical evaluation of the proposed methods and demonstrated that semantic aspect-oriented representations improve both predictive performance and explainability, fostering biomedical research. This work further highlights that in complex and multi-disciplinary domains, where a single knowledge graph is used to support a wide variety of tasks, it is essential to shift from viewing knowledge graphs as a whole to focusing on specific semantic perspectives.application/pdfpt_PTSemantic perspectives for learning over biomedical knowledge graphsSousa, Rita Isabel Torres dePesquita, Cátia Luísa Santana CalistoSilva, Sara Guilherme Oliveira daHostingInstitutionOrganizationalRepositório Científico de Acesso Aberto da ULisboae-mailmailto:repositorio@reitoria.ulisboa.ptrepositorio@reitoria.ulisboa.ptURNurn:tid:1016675742025-02-18T15:39:03Z2024-07-012023-12-292024-07-01T00:00:00ZHandlehttp://hdl.handle.net/10400.5/98530http://purl.org/coar/access_right/c_abf2open accessKnowledge graphSemantic similarityKnowledge graph embeddingMachine learningBiomedical applicationGrafo de conhecimentoSemelhança semânticaEmbedding de grafos de conhecimentoAprendizagem automáticaAplicação biomédica25679013 bytesFundação para a Ciência e a TecnologiaNovo: Semantic perspectives for learning over biomedical knowledge graphs. Inicial: Evolving meaning for supervised learning in complex biomedical domains using knowledge graphsCrossref Funder IDhttp://doi.org/10.13039/501100001871literaturehttp://purl.org/coar/resource_type/c_db06doctoral thesishttp://purl.org/coar/access_right/c_abf2application/pdffulltexthttps://repositorio.ulisboa.pt/bitstreams/9daa1ab8-c388-436e-ad8d-00927f30e953/download |
| spellingShingle | Semantic perspectives for learning over biomedical knowledge graphs Sousa, Rita Isabel Torres de Knowledge graph Semantic similarity Knowledge graph embedding Machine learning Biomedical application Grafo de conhecimento Semelhança semântica Embedding de grafos de conhecimento Aprendizagem automática Aplicação biomédica |
| status | SINGLETON |
| subject.fl_str_mv | Knowledge graph Semantic similarity Knowledge graph embedding Machine learning Biomedical application Grafo de conhecimento Semelhança semântica Embedding de grafos de conhecimento Aprendizagem automática Aplicação biomédica |
| title | Semantic perspectives for learning over biomedical knowledge graphs |
| title_full | Semantic perspectives for learning over biomedical knowledge graphs |
| title_fullStr | Semantic perspectives for learning over biomedical knowledge graphs |
| title_full_unstemmed | Semantic perspectives for learning over biomedical knowledge graphs |
| title_short | Semantic perspectives for learning over biomedical knowledge graphs |
| title_sort | Semantic perspectives for learning over biomedical knowledge graphs |
| topic | Knowledge graph Semantic similarity Knowledge graph embedding Machine learning Biomedical application Grafo de conhecimento Semelhança semântica Embedding de grafos de conhecimento Aprendizagem automática Aplicação biomédica |
| topic_facet | Knowledge graph Semantic similarity Knowledge graph embedding Machine learning Biomedical application Grafo de conhecimento Semelhança semântica Embedding de grafos de conhecimento Aprendizagem automática Aplicação biomédica |
| url | http://hdl.handle.net/10400.5/98530 |
| visible | 1 |