Fuzzy Ontology Based Multi-Modal Semantic Information Retrieval

Lack of today's technology of information retrieval system is the semantic between the information resources. The semantic can be expressed for a resource using the underlying technology of ontology. So the main objective of this paper is to design an information retrieval system for both text and image data using the concept of ontology. The main focus of this paper is to improve information retrieval for sports events using Ontologies. As it is complex problem, it is addressed into the following sub-problems in this paper. (1) Integrating domain knowledge and images using fuzzy ontology and Retrieving the required Muti-modal information using fuzzy rule set. (2) Providing image semantic by constructing visual codebook for affine covariant-Semantic segmented patches. (3) Analysing the discriminative power of each visual word using the probabilistic latent semantic and quantizing them using the Chi-Square test. In this work, the domain of Sports is considered for creating both Low-level visual ontology for certain sport event images and also for building a high-level domain ontology from the information on web. These two Ontologies are integrated using Fuzzy concepts.