Design And Implementation Of Context Reasoning And Computing Subsystem Based On Ontology

With the continuous development of mobile networks and the capabilities of mobile intelligent terminal device continue to increase, making it possible to obtain context information. Context information is any information that characterizes the situation of an entity. By analyzing the context environment where the user located, to provide user with personalized service, is a hot topic in the mobile domain.The author participates in the selection of personalized mobile internet service system, which considers the current context as an important factor impacting user in choice of service. Through the analysis of context environment to provide more accurate, personalized choice of service for user. The design and implementation of context reasoning and computing subsystem based on ontology is a subsystem of the selection of personalized mobile internet service system. The subsystem includes the context acquisition, context processing, context reasoning and context calculation four functions. The subsystem provides data source, context query interface and context calculation interface for the user preference extraction subsystem, service selection subsystem, the user preference of the prototype system.Firstly, this paper analyzes the context information in mobile networks, which be divided into two types of low context and high context, and gives some methods of how to acquisition these low context. Secondly, through pretreatment of context information to construct a domain ontology model for mobile, which supports of context sharing and reasoning. Thirdly, based on the ontology model, through customizing inference rules, using Jena inference engine reasoning high level context from low level context, which riches the types of context information. Fourthly,based on the concept hierarchy and attributes of ontology model, focus on the design of context similarity algorithms, including the semantic similarity calculation and description similarity calculation. On this basis, by calculating the similarity between the current context and the historical context and getting a service matching set which associates with the current context, so that it can be applied to service selection subsystem. Finally, test the system with unit testing and integration testing.