Author
Elena Camossi, Elisa Bertino, Giovanna Guerrini, Michela Bertolotto
Abstract
In applications involving spatio-temporal modelling, granularities of data may have to adapt according to the evolving semantics and significance of data. To address such a problem, in this paper we define ST2_ODMGe, a multigranular spatio-temporal model supporting evolutions, which encompass the dynamic adaptation of attribute granularities,
and the deletion of attribute values. Evolutions are specified
as Event - Condition - Action rules and are performed at run-time. The event, the condition and the action may refer to a period of time and a geographical area. Periodic evolutions may be specified, referring to both transaction and valid time dimensions. The evolution may also be constrained
by the attribute values. Evolutions greatly enhance
exibility in multigranular spatio-temporal data handling but require revisiting the notion of object consistency with respect to class definitions and access
to multigranular object values.