A framework for integrating DBpedia into a multi-modality ontology image retrieval
DBpedia, used as a web link knowledge garden, provides great opportunities for researchers as a domain concept to enrich resource and information extraction. The integration of DBpedia with ontology-based approach in image retrieval gives complete and rich semantics information to the image. The sem...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IJoAT Foundation
2013
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/33073/ http://irep.iium.edu.my/33073/ http://irep.iium.edu.my/33073/3/IJACT3188PPL%5B1%5D.pdf |
| Summary: | DBpedia, used as a web link knowledge garden, provides great opportunities for researchers as a domain concept to enrich resource and information extraction. The integration of DBpedia with ontology-based approach in image retrieval gives complete and rich semantics information to the image. The semantic gap is the main problem in image retrieval. The gap is between the high level image interpretations of the users and the low level image features stored for indexing querying. Ontology-based image retrieval is an effective approach to bridge the semantic gap because it is more focused on capturing and presenting the semantic content which has the potential to satisfy the user need. A recent trend in ontology-based image retrieval is to fuse the two basic modalities of images namely textual content (keywords) and visual features and known as multi-modality ontology. In this paper, we present the framework for integrating structured content in DBpedia resources with multi- modality ontology-based image extraction and retrieval system and describe how this framework bridges the semantic gap in content-based image retrieval (CBIR). Our goal is to populate a knowledge base with online image news resources from 12 sport types in the BBC sport news, which has three main items: image, image caption and news information. This system will yield high precision and include diverse sports images for specific entities. A multi-modality ontology retrieval system, with complete relational facts about entities will improves the precision of retrieval. |
|---|