The Study And Design Of Personalized Learning System Based On Ontology

With the development and popularization of network technology, the network has become users’main way of accessing information, which also is one of the most convenient ways. The fast spread of network information provides a more convenient way for individuals and society’s development, points out a new direction for the evolving new education pattern, which kicks off an age of network education.Although online learning system breaks the time and space limit of traditional teaching mode to a certain extent, but the traditional online learning system teaches fixed content with a single teaching mode, which helps students achieve self-learning not so well. Therefore, the personalized learning system came into being, which can not only provide learning resources according to the student’s clear requirements, but also meet the personalized learning needs of students, truly individualized.In this paper, we analyze in-depth the related theory and technology of personalized learning and construct a personalized learning system based on ontology. This system can be divided into five main modules, the ontology-based semantic similarity management module, resource management module, student learning management module, personalized recommendation module and system management module. The ontology-based semantic similarity management module lays the foundation for accurately constructing text vector model of learning resources as well as building and updating user interest model, mainly through establishing course ontology and then on this basis analyzing concept attribute, concept instance, the relationship between the conceptual structure and conceptual name, and mining the semantic similarity between concepts. Resource management module mainly realizes the management and maintenance of learning resource as well as the text model building, updating, etc. This module automatically process each learning text uploaded by teacher users and construct a text vector model based on concept semantic similarity relation for personalized recommendation module to use. Student learning management module includes student interest model building, updating, and student learning style choosing, and other functions. Due to the accuracy and timeliness of user interest model directly determines the performance of the system is good or bad, so we make full use of the concept semantic similarity information mined based on ontology and extend related concepts on user interest model in semantic level, thereby more accurately describe the student interest model and improve the precision of text recommendation. Personalized recommendation module compares the similarity of students’ interest model and text vector model of learning resources, and then recommends higher-similarity learning resources to users.At last, we compares with an experiment recommended effects generated by traditional text vector modeling method&user interest modeling method and ontology-based text vector modeling method&user interest modeling method, proved that the modeling method proposed in this paper improves the accuracy of text recommendation to a certain extent.