Bees Swarm Optimization for Real Time Ontology Based Information Retrieval

In this paper, we propose an extended IR vector space model by introducing some semantics concepts inherent to YAGO ontology (A large ontology derived from Wikipedia and Word Net), then an adaptation of Bees Swarm Optimization algorithm (BSO) that was developed previously for classical IR, to the new context. In the offline indexing step we define a representation based on semantics annotations taken from YAGO ontology, for the documents and the queries. Afterwards, we integrate the appropriate structures obtained at the first stage in the online interrogation process. The experimental tests have been performed on Reuter´s corpus and compared to those previously obtained. The preliminary results clearly show that this pathway promises to provide efficient results for real time IR and deserve to be further deepened.