A Dynamic Dominance-Based Rough Set Approach for Processing Ordered Data

In real-world many information systems are varying over time. How to process efficiently dynamic data is a hot issue in data mining research field. Processing information with preference-ordered attribute domain is an important task in multi-criteria decision making. Since dominance-based rough set approach can process information with preference-ordered attribute domain, it has been applied widely to multi-criteria decision making. In the paper, we introduce a new strategy of updating approximations in DRSA under the cases when a lot of objects being added. For thus, the notation of base boundaries of approximations in DRSA is given to reduce the redundant computation in updating approximations. The feasibility of the approach was validated by a numerical example.