Semantic enrichment of sensor data: a case study in environmental health

• Ano de defesa: 2021
• Autor(a) principal: Silva, Lucas Felipe Moreira
• Orientador(a): Bulcão Neto, Renato de Freitas
• Banca de defesa: Bulcão Neto, Renato de Freitas, Macedo, Alessandra Alaniz, Sene Júnior, Iwens Gervásio
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Goiás
• Programa de Pós-Graduação: Programa de Pós-graduação em Ciência da Computação (INF)
• Departamento: Instituto de Informática - INF (RG)
• País: Brasil
• Palavras-chave em Português: Engenharia de software; Informática em saúde; Web semântica; Ontologia; IoT-Stream; Fluxos de dados; Enriquecimento semântico; Saúde ambiental
• Palavras-chave em Inglês: Software engineering; Health informatics; Web semantic; Ontology; IoT-Stream; Data streams; Semantic enrichment and environmental health
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::SISTEMAS DE COMPUTACAO
• Link de acesso: http://repositorio.bc.ufg.br/tede/handle/tede/11631

Indoor Air Quality is crucial for human health, but over ninety percent of people worldwide breathe air with pollutant levels that exceed the WHO limits, which may trigger or worsen symptoms the longer one stays exposed. Studies in the area face an inherent difficulty: the massive number of interconnected elements and the effects on human health. However, IoT technologies like sensors and actuators are helping the field address this problem by acquiring and processing EH data to be used in automation and decision-making. Still, although sensors deployment is relatively simple and feasible, raw data is barely useless in practice, requiring preprocessing before usage and is highly dynamic, meaning sensor data for Environmental Health (EH) should be handled as data streams. Streams can be enriched with information such as air quality indexes and associated with curated medical knowledge, improving usage. IoT's regular data life cycle comprises acquisition, modeling, reasoning, and distribution, so a first step to enable an IoT-based EH scenario is a shared common representation for EH data acquired from sensors, which can be met by Ontologies' expressiveness and reasoning support. The organization of the fundamental processes of IoT-based EH systems into a reference architecture can further support the development of such systems and a Reference Architecture like RAISE, whose central idea is to structure general software components into a reusable design solution for semantic enrichment of healthcare data attain this task. That process comprises steps like acquisition, modeling, extraction, preprocessing, semantic annotation, integration, and storage of heterogeneous healthcare information. The problem addressed here is the low number of validation research investigating semantic enrichment and integration of EH data through ontologies and medical knowledge. This work's objective was to elaborate on an instance of the RAISE architecture that enriches sensor data for the EH domain, contributing with: Semantic Enrichment of EH sensor-acquired data; The link between ontologies to address the complete picture; and more.