Vulcont: A Recommender System based on Contexts History Ontology

Detalhes bibliográficos
Ano de defesa: 2017
Autor(a) principal: Cardoso, Ismael Messias Gomes
Orientador(a): Barbosa, Jorge Luis Victória
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Universidade do Vale do Rio dos Sinos
Programa de Pós-Graduação: Programa de Pós-Graduação em Computação Aplicada
Departamento: Escola Politécnica
País: Brasil
Palavras-chave em Português:
Palavras-chave em Inglês:
Área do conhecimento CNPq:
Link de acesso: http://www.repositorio.jesuita.org.br/handle/UNISINOS/6352
Resumo: The use of recommender systems is already widespread. Everyday people are exposed to different items’ offering that infer their interest and anticipate decisions. The context information (such as location, goals, and entities around a context) plays a key role in the recommendation’s accuracy. Extending contexts snapshots into contexts histories enables that information to be exploit. It is possible to identify context’s sequences, similar contexts histories and even predict future contexts. In this work we present Vulcont, a recommender system based on a contexts history ontology. Vulcont merges the benefits of ontology reasoning with contexts histories in order to measure contexts history similarity, based on semantic and ontology’s properties provided by context’s domain. Vulcont considers synonymous and classes’ relations to measure similarity. After that, a collaborative filtering approach identifies sequences’ frequency to identify potential items for recommendation. We evaluated and discussed the Vulcont’s recommendation in four scenarios in an offline experiment, which presents Vulcont’s recommendation power, due the exploit of semantic value of contexts history.