Research On Context Awareness In Mobile Environment

With the rapid development of mobile environment,users are increasingly dependent on mobile devices,and the requirements for mobile services are getting higher and higher.Coarse-grained service model is no longer suitable for the rapid development of information society,the pursuit of information services personalized and delicate.At the same time,mobile devices are widely used,but also produce massive context data information.How to dig out the value of mobile information,how to collect the information used to enhance the service,which produced the research motivation.Context-aware technology is a research about using collected context data to enhance service.In mobile environment,devices can generate and collect large amounts of context information to meet the needs of context-aware applications.How to make full use of context-aware technology in the mobile environment to improve the quality of information service is the core of this research.In this thesis,through the research,we hope to solve several problems,that is,how to accurately express the complex mobile context,how the overall process of context-aware in the mobile environment works,how to improve the satisfaction of mobile context-aware service,how to use context-aware to solve the traditional recommendation problem.In view of the above problems,in this thesis,we first define the whole process of mobile context-aware,introduce the idea of context classification and the way of context reasoning.Then consider the ternary relationship about resource-user-context,to the user as the center,and use the ontology and Protégé tools to make a complete modeling of the mobile context.In order to analyze the mobile context-aware for the promotion of information services,choose to use the recommended system for verification.First analyze the impact of the mobile environment on the recommended algorithm,Secondly,by improving the traditional formula,add context information into collaborative filtering.And then put forward own solutions to the problems of traditional recommendation.In order to solve the problem of cold start,the factors influencing the interest and the social factors are analyzed.The problem of data sparsening is solved by the generalization of the situation.Then through the matching algorithm,combined with the ontology model,further using of context information,merge the context pre-filtering and modeling of the hybrid mechanism of the recommended algorithm.Finally,through the design of the mobile context recommendation system,combined with the corresponding indicators of the test,the use of the system verifies the feasibility of mobile context into the algorithm,further proving the effectiveness of context-aware in the mobile field.Finally,the research is summarized and forecasted,and the shortcomings of this thesis are discussed,and the future research direction is put forward.