Utilize este identificador para referenciar este registo: http://hdl.handle.net/10362/179727
Registo completo
Campo DCValorIdioma
dc.contributor.advisorHan, Qiwei-
dc.contributor.authorTheilmann, Alex-
dc.date.accessioned2025-02-25T09:45:30Z-
dc.date.available2025-02-25T09:45:30Z-
dc.date.issued2024-01-19-
dc.date.submitted2023-12-20-
dc.identifier.urihttp://hdl.handle.net/10362/179727-
dc.description.abstractThis paper explores entity matching and its vital role in e-commerce to track products across different domains. Focusing on five diverse approaches, we evaluate their performance based on the precision to recall trade-off. Each model is examined theoretically, emphasizing its advantages and limitations. We also discuss potential improvements to enhance their applicability in this field. Our research aims to provide deeper insights into the effectiveness of various entity matching strategies and their performance under distinct priorities, while considering model architecture and room for improvement.pt_PT
dc.language.isoengpt_PT
dc.relationUID/ECO/00124/2013pt_PT
dc.rightsopenAccesspt_PT
dc.subjectEntity matchingpt_PT
dc.subjectRecord linkagept_PT
dc.subjectEntity resolutionpt_PT
dc.subjectData linkagept_PT
dc.subjectE-Commercept_PT
dc.subjectMachine learningpt_PT
dc.subjectNIPpt_PT
dc.subjectCommunity detectionpt_PT
dc.titleExploring product entity matching in a multi-domain landscape - a community detection approachpt_PT
dc.typemasterThesispt_PT
thesis.degree.nameA Work Project, presented as part of the requirements for the Award of a Master’s degree in Business Analytics from the Nova School of Business and Economicspt_PT
dc.identifier.tid203867394pt_PT
dc.subject.fosDomínio/Área Científica::Ciências Sociais::Economia e Gestãopt_PT
Aparece nas colecções:NSBE: Nova SBE - MA Dissertations

Ficheiros deste registo:
Ficheiro Descrição TamanhoFormato 
2023_2024_Fall_53679.pdf1,8 MBAdobe PDFVer/Abrir


FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpace
Formato BibTex MendeleyEndnote 

Todos os registos no repositório estão protegidos por leis de copyright, com todos os direitos reservados.