【正文】 O is a resource, then O is the center of expansion, the concepts for a query can be obtained according to the concepts and attribute relations of O. The query processing is the most important part of semantics retrieval. It can be described as the following RI model: ontology model and ontologybased web resources identity model. In those, web resources R and the query Q are built based on ontology o and they are relevant if and only if R satisfy Q. It describes that R and O imply logically the query Q: O^RQ [9]. Fig 2. Processing of user request There are two proven methods that can convert the query keyword into ontology description. [10] proposed a semanticbased retrieval approach: ontology matching graph, that is, semantic correlations are brought forward between ontology, relationships and ontology graph. According to these semantic correlations, the framework for ontology graph matching that can pute these correlativities is designed. In [11], an alternative idea of semantic retrieval is given. It presents a search architecture that bines classical search techniques with spread activation techniques applied to a semantic model of a given domain. Given ontology, weights are assigned to links based on certain properties of the ontology, so that they measure the strength of the relation. Spread activation techniques are used to find related concepts in the ontology given an initial set of concepts and corresponding initial activation values. These initial values are obtained from the results of classical search applied to the data associated with the concepts in the ontology. Two test cases were implemented, with very positive results. It was also observed that the proposed hybrid spread activation, bining the symbolic and the subsymbolic approaches, achieved better results when pared to each of the approaches alone[11]. D. Inference engine This is the critical control module of the whole framework. Through the inference and search, knowledge user’ need will be obtained, at the same time, intelligent teaching methods and strategies can be gotten. Inference engine is responsible for parsing and reasoning owl documents. The aim is to read ontology from general file, store in a specific model in order to facilitate process, and then carry on ontologybased semantic inference according to certain rules, which is the semantic retrieval crucial step. This process uses Jena development kit to implement. As shown in Figure 3, the operational principle of inference engine is: inference engine registration mechanism creates reasoning machine based on basic RDF triples and ontology. After this process, model objects contained inference mechanism, InefreneeGnarh and InGfraph can be generated. In Jena, Garph is also known as Model and its manifestation is ModelInterafee. And then, it can take on operation and processing to this model by using the Model API and Onotlogy API, information retrieval on semantic level can be achieved consequently. Fig 3. Operational principle of Inference Engine V. CONCLUSION AND FUTURE WORKS In this paper, we have proposed the design, implementation of a new application framework for work education resources library. By using ontologybased semantic technology, we create an open platform for new educational resources, achieve interoperation between heterogeneous resources and realize the centralized management of resources and contents distributed stored. Based on these, the resource library can bee sharing, distribution, open and make interoperation between heterogeneous resource be supported. With the rapid development of educational contents, curriculum is also changing. The production and update of education resources are more frequently. Through using ontology’ standardized expression and work process and establishing education resource library based on mon standard, the widespread sharing and reuse of resources can be achieved and the consensus of concept can be formed. It will be an important direction for construction and development of education resource library. In the future work, we will extend our current work along three directions. First one is the improvement of the architecture. Second direction is to realize it with more plex ontologies when rich data sets related to our framework are available. The last is paring with other work education resources library. ACKNOWLEDGMENT This work is supported by the National Natural Science Foundation of China (NSFC) under Grant and the Specialized Research Fund for the Doctoral Program of Higher Education of China under Grant No. 20200183044.