Evolution Specification of Multigranular Temporal Objects

The main key feature of temporal databases is to maintain all values taken by object attributes over time. Since historical information may be needed at different levels of detail, multigranular temporal databases have been introduced, in which different attributes can be stored at different temporal granularities. An important issue that has not been addressed, however is that the required level of detail does not only depend on attribute semantics, rather it is often related to how recent the data are. It is quite natural that recent data are needed at greater level of detail, whereas less detail is needed as data age. As an extreme case, data can also expire, that is, they are no longer needed after a certain period of time. In this paper we address the problem of evolution and expiration of historical data in a multigranular temporal object model.

Authors

• Elena Camossi
• Elisa Bertino
• Giovanna Guerrini
• Marco Mesiti

Keywords

• Temperature
• Aggregates
• Meteorology
• Spatial databases
• Weather forecasting
• Cities and towns
• Banking
• Pollution
• Kernel
• Power generation
• Objective Specification
• Different Levels Of Detail
• Temporal Granularity
• Greater Level Of Detail
• Set Of Values
• Set Of Types
• Development Of Conditions
• Aggregate Level
• Types Of Values
• Temporal Data
• Remainder Of This Section
• Number Of Months
• Data Warehouse
• Aggregation Function
• Temporal Interval
• Attribute Values
• Evolution Operator
• Finer Granularity
• Running Example
• Temporal Elements
• Legal Value
• Database Schema
• Static Type
• Tax Payments
• User-defined Function
• Temporal Dimension
• Class Instances
• Temporal Window