Research On Processing Method Of Incomplete Information System Based On Rough Set

This paper uses the rough set theory as tool, incomplete information system as object, acquiring knowledge as purpose, for studying method which is based on rough set theory in incomplete information to acquire knowledge. Specific contents include learning extended rough set model, filling missing data, and reduction of attributes in incomplete information system. The main contents are summarized as follows:First, this paper describes the current status of research which to fill missing data in the incomplete information system, analyzing the ROUSTIDA algorithm which is better to fill. On this basis, this paper puts forward a filling method which is based on rough set Biclustering(MFBIB) about missing data. This method fills missing data based on the theory that the mean squared residue of Biclustering perfect cluster is 0, and the volatility of the cluster’s attribute values is consistent. This paper translates the problem of finding the maximum perfect cluster which contains the missing values into the problem of finding out the maximum similarity attribute sets between the missing object and other objects through mathematical analysis, then the majority of missing values which is used as the filling value can be calculated out by the same maximum similarity attribute sets. This paper takes experiments using four groups of UCI data sets, and it is demonstrated that MFBIB averagely improves the accuracy of 77.13% filling values compare to ROUSTIDA.Second, this paper describes two common reduction of attributes in incomplete information system, it is based on distinguish matrix and entropy. After summarizing the characteristics of two algorithms, combining MFBIB algorithm, this paper puts forward a algorithm about reduction of attributes which is based on tree structure. This algorithm uses tree structure to store object attributes, and using Biclustering perfect cluster to distinguish compatibility of between objects. Than to identify the core attributes about incomplete information system to get the reduction of attributes. At last, we use case to analyze to verify the feasibility of this algorithm.