Publicação

Dados.IPB: Making Research Data Discoverable, Interoperable and Reusable

Ver documento

Detalhes bibliográficos
Resumo:Data is a crucial resource in the Knowledge era, and data repositories are important tools for sharing and preserving it. In open science, it is mandatory to make data repositories easily discoverable and usable across tools to generate big and complex datasets - Big Data. Machine Learning (ML) uncovers patterns in Big Data and enables, for example, automatic object detection in images for fraud detection. IPB maintains its Research Data Repository (DADOS.IPB1), aggregated with other data repositories. DADOS.IPB metadata are integrated with metadata from other OpenAIRE repositories (via the OAI-PMH protocol). Using a Python script to collect all Croissant metadata files and their respective files from all Dataverse instances worldwide. This article presents a case study of detecting Bactrocera Oleae exemplars in photos using YOLO, achieving high accuracy.
Autores principais:Alves, Adília
Outros Autores:Pais, Clarisse
Assunto:Dados.IPB Open ccience Metadata Croissant metadata format Machine learning
Ano:2026
País:Portugal
Tipo de documento:documento de conferência
Tipo de acesso:acesso aberto
Instituição associada:Instituto Politécnico de Bragança
Idioma:inglês
Origem:Biblioteca Digital do IPB
Descrição
Resumo:Data is a crucial resource in the Knowledge era, and data repositories are important tools for sharing and preserving it. In open science, it is mandatory to make data repositories easily discoverable and usable across tools to generate big and complex datasets - Big Data. Machine Learning (ML) uncovers patterns in Big Data and enables, for example, automatic object detection in images for fraud detection. IPB maintains its Research Data Repository (DADOS.IPB1), aggregated with other data repositories. DADOS.IPB metadata are integrated with metadata from other OpenAIRE repositories (via the OAI-PMH protocol). Using a Python script to collect all Croissant metadata files and their respective files from all Dataverse instances worldwide. This article presents a case study of detecting Bactrocera Oleae exemplars in photos using YOLO, achieving high accuracy.