Zz-structures are particular data structures capable of representing both hypertext data information and contextual interconnections among various chunks of information. We consider an extension of the standard zz-structure model in terms of computational agents. In particular, we propose a multi-agent adaptive system in which users store their personal information inside a zz-structure and agents collaborate in order to support them in the extraction of personalized navigational views, allowing creation of personalized tours, for example tours within virtual museums. The strength of this new model resides in the level of freedom users have for the dynamical choice, based on some present interest or necessity, of their navigational path inside the virtual museum.
