The Study Of Citrus Fertilizer Decision-Making System Based On Semantic Technology

China is the main citrus producing country. Traditional citrus production management has some weak points, such as over-reliance on "experience", high demand for labor, and requirement of experienced workers. It is difficult to approach large-scale, real-time production management. It results in insurmountable gap between china and other citrus developed countries like United States and Brazil, In order to solve the current problems, the government proposes an approach based on agricultural information system to promote the development of modern agriculture. However, how the massive, multi-source, heterogeneous agricultural domain knowledge converted into practice is an urgent problem in agricultural information system. What people used to do developed expertise knowledge system based on relational database. But, there were a lot of shortcomings in reality, such as query inaccuracy, less knowledge, cannot handling multi-dimensional problem etc. Ontology based on semantic technology is crucial to help solve these issues. Because, semantic ontology, which use Uniform Resource Identifier (URI) and W3C’s standard language such as RDF, RDFS, RDFS+and OWL, has advantages on semantic integration and large-scale reasoning. Therefore, semantic technology provides the possibility of applying agriculture knowledge to practical life.By studying the semantic technology, we propose an approach based on Semantic Web to build citrus fertilizer ontology which solve agricultural information system encountered problems. The main focuses of this paper are as follows:Firstly, we have integrated massive, multi-source, heterogeneous citrus production management information through semantic way. According to the rules of building ontology, we build citrus fertilizer ontology through several steps, such as knowledge extraction, classification, definition, building triples, building rules, query validation and so on.Secondly, we have found redundancy problem in semantic reasoning. In this paper, we choose AGROVOC ontology as example which is built by FAO and concludes agriculture knowledge, forestry knowledge, fishery knowledge and other related fields to explore rationality of building ontology form the perspective of inference redundancy. This paper therefore employs inference mechanism and the methods of complex network analysis to analyze and prove inference redundancy. The final results showed that it provides a new angle and method for the examination of ontology reasoning rationality.Thirdly, we have developed decision-making system based on Semantic Web to provide proactive, personalized, precise citrus production management information services. Ordinary farmers are lack of expertise in agricultural production (such as irrigation, fertilization, pest, etc.), so we combine with citrus fertilizer ontology to establish citrus fertilizer and disease query system. It is greatly helpful to resolve unavailability of rural information service issues especially the problem applying agriculture knowledge to practical life.Finally, this paper propose an approach based on Semantic technology to build ontology and decision-making system on citrus fertilizer for agriculture informatization. We believe that this work will provide some beneficial reference for solving similar problems in future.