A Real Rough Set Model Based on Significance and Its Applications to Traffic Condition Recognition

The notion of a new real attribute significance is introduced by analyzing the limitations of the current theories and algorithms about the rough sets applied to real decision system. Then a real rough set model and the fast attribute reduction algorithm are presented based on the presented notion, which can avoid data discretization in traditional rough set theories. Finally, the presented algorithm and several other algorithms are applied to region traffic condition recognition, and the comparison results show that the proposed algorithm is of high classification accuracy.