Knowledge Based On Information Entropy In Rough Set Reduction Algorithm Research

Simplification of knowledge is deeply studied,which is one of the core problems in Rough Set data mining.In the thesis,the knowledge reduction is studied from the viewpoint of information theory.Firstly,the algebra express and information express of Rough Set theory are analyzed and compared with each other systematically.Some laws are discovered as following:1)When the number of the conditional attributes is increasing,the changing tendency of the conditional entropy of decision attributes for condition attributes is non-rigorous monotoni -cally decreasing;2)Suppose the reduction process starts from the core of a decision table, when an un-removable attribute being added to the reduction,the conditional entropy of decision attributes for the reduction is monotonically decreasing;3)The conditional entro -py of decision table will not change in the reduction process.Then,a heuristic algorithm based on the intrinsic relation between the information table and decision table to reduce decision table is proposed in this thesis.Secondly,a hierarchical reduction algorithm of Rough Set theory is proposed.Based on the information theory,it is proved that the entropy of information system and the mutual information of decision system are constant in the hierarchization of attributes.So the Rough Set theory hierarchical reduction algorithms have strict mathematic basis.An improved hierarchical reduction algorithm based on the significance of attributes is proposed in this thesis.Finally,after analyze the attribution reduction algorithm based on Rough Set that has arisen at present,a new and relatively reasonable formula measuring the significance of attributes is given,and the property of this formula is analyzed.Then,a quick attribute reduction algorithm is proposed.