Semantic Retrieval System Based on Corn Ontology

Most of the current information retrieval systems are mainly based on full text matching of keywords or topic-based classification, often return a large number of irrelevant information, and are unable to meet the user´s request. Ontology-based semantic retrieval is a hot issue in current research. In this paper, the corn plant ontology is constructed using Formal Concept Analysis based approach in which the concept lattice is built from terminology-file relationship table and further reduced. Based on the corn plant ontology, we propose a semantic annotation method in which the feature words are selected by an improved method for weight calculation and the RDF triples are generated by syntactic parser. Finally a semantic retrieval system for corn plant is developed. In comparative experiment one hundred documents are selected as the dataset, and the result shows that the semantic retrieval system introduced in this paper is superior to keyword-based retrieval method in precision ratio and recall ratio.