An Information Retrieval Application using Ontologies

Christian Paz-TrilloRenata WassermannPaula P. Braga

Searching for information in long videos can be a time-consuming experience. In this paper, we describe OnAIR, an ontology-aided information retrieval system applied to retrieve clips from video collections.We used a video collection compiled frominterviews with Ana Teixeira, a Brazilian artist. Theinterviews were made by Paula P. Braga, the domainexpert. The interview is developed in the domain ofcontemporary art and the system uses a domain ontologyto expand the queries with related terms. We tested thesystem with a battery of queries, and we veri.ed that theontology contributes to the e.ciency improvement in termsof the relevance of retrieved documents. We designed thesystem to work in a domain-independent way, allowingus to move to other domains by just changing theunderlying ontologies and video collections.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: