An efficient algorithm for rough sets positive region

In rough sets theory, attribute reduction is considered as an important preprocessing step for machine learning, pattern recognition, and data mining. The algorithms of attribute reduction and attribute core on rough sets are main content of rough sets theory. Positive region algorithm is an important branch. Many positive region algorithms have been proposed, however the time and space complexity is relatively high. To overcome this shortcoming, we introduce the size of positive region algorithm and positive region algorithm based on the simplification decision table. In order to verify the efficiency of the algorithms, we design several efficient relative core algorithms. Experiments show the proposed methods have lower time complexity and space complexity. It is worth noting that the improvement becomes more profoundly visible when dealing with larger data sets.