Abstracto

SEMANTIC RETRIEVAL TECHNIQUE BASED ON DOMAIN ONTOLOGY

Ms.Amrata Soni, Ms.Hemlata Sunhare, Mr.Sachin Patel

The Semantic Web is a latest technique for organizing information and it characterize a huge interest area for the worldwide research group of people, however it is still far from a significant implementation. In this research we have proposed a scheme for information retrieval stand on ontologies, dynamic semantic network, and lexical chains, significant an approach for achieve and indexing consequences by means of a narrative metric to compute semantic relatedness between words. Our technique has numerous novelties, in meticulous concerning the utilize of a collective knowledge base from which we mine precise domain ontologies; furthermore, the anticipated semantic relatedness metric achieve optimally evaluate with further metrics in the using a general test set. In meticulous we are taking into consideration the opportunity of introduce several form of normalization for the semantic constituent with respect to the length of lexical chains and the size of the documents, applying our research news paper domain and we are improving the analysis accuracy of our system compared with other metrics.