Human can increase the conversational bandwidth using implicit situational and environmental information. But it is dfjcult to apply implicit information to current data model in order to enhance the communication a b i l i ~ between human and computer. In this paper, we are interested in data modeling of context information required to support 4namic sewice adaptation for increasing communication ability in user-centric environments. By improving the access capacity to context information for sewice adaptation, we can increase the richness of communication in human computer interactions and make it possible to produce more useful computational sewices. We are interested in the structure of meta-data model of context information required to support dynamic sewice adaptation in user-centric environments. Users want to get sewices anytime, anywhere and on any devices in various 4namic environments. The meta-data of context information is important for offering dynamic sewices and the reconjguration context information for dynamic sewice adaptation. This paper designs a structure of nzeta-data nzodel that supports the reconjguration of the context for 4namic sewice adaptation. We classzJJ the context information, as well as dejne its relationship with other context information on the basis of the application sewices, changes in the external environments.