National Agricultural Break This Collaborative Construction Of A Number Of Semantic Retrieval Technology Research

Ontology is an important emerging discipline that has huge potential to improve information organization, management and understanding. It has a crucial role to play in enabling content-based access, interoperability, communications, and providing qualitatively new level of services on Semantic Web. In this paper, the main work focuses on the collaborative construction of agricultural ontology and sematic retrieval for agricultural resources.Firstly, the agricultural data are massive, professional and the domain experts usually reside in different region, which result in the isolated information that is difficult for sharing. It is a hotbed for collaborative conflits that are difficult to detect in the process of collaborative ontology construction. Therefore, we propose a method to classify and detect collaborative conflicts with solution to resolve conflits when they are detected.Secondly, there is great deal of unstructured documents of agricultural resources. It is hardly to transfer them into formal knowledge expressed by ontology. Therefore, we propose a semantic annotation method to annotate them by agricultural ontology. The method annotates document by two steps, one is to classify documents by feature-based Navive Bayes automatically, and the other is to calculate the correlation between ontology entity and document by the semantic context of the entitiy and the structural infomation of the document.Thirdly, a semantic retrieval model is proposed based on the domain knowledge of agriculture. This model provides an interface for user to input natural language, then it maps identified keywords to ontology entities by calculating the correlation and similarity between them; furthermore, we give a set of rules for semantic extension based on the features of agricultural ontology; and we combine the result got from the ontology with the annotated documents.Lastly, we present a management and service system for agricultural knowledge and resources. It has two parts, one is the collaborative ontology construction module, and the other is semantic retrieval module based on agricultural ontology. The former is a tool for constructing agricultural ontology collaboratively, providing functions for detecting and resolving conflicts. The latter is a semantic retrieval tool that can transfer the users' natural language into formal search and return the result that combines the knowledge expressed in ontology with annotated documents to users.