Comparative Research of Attribute Reductions based on the System Entropy and on the Database System

Attribute reduction is one of important research in the rough set theory. The relationship of many deferent attribute reduction definitions called relative attribute reductions were researched by some researchers. But the rules acquired from relative attribute reduction have some shortcomings. To overcome these drawbacks, some researchers proposed other type attribute definitions called attribute reduction definition based on the system entropy and on the database system. The relationship of above two new attribute reductions was researched, and it is proved that the two new attribute reductions are equivalent to each other. This consult can help us design efficient algorithm of attribute reduction and algorithm of rules acquired form decision table based on system entropy or on database system with more ways.