Papers by Keyword: Uncertain Information System

Abstract: Based on knowledge equivalent classification of an uncertain system, from view of the knowledge coordinating relations and dependence of condition attributes and the decision attributes in an information universe，a new concept of information consistency of a database is presented to simplify knowledge of an uncertain system. A basic algorithm and realization of knowledge simplification on the information consistency of a database is suggested in this paper. It is more legible and convenient for knowledge simplification by way of the information consistency of a database than the idea of knowledge dependence. The feasibility of the proposed approaches of knowledge simplification and reasoning decision rules is validated by some of example of a classic CTR car knowledge representation system with a large-scale database here.

Abstract: To attribute reduction in an uncertain information system, this paper proposed a method of attribute reduction based on rough set theory. This reduction method gives the concepts of tolerance relationship, attribute significance and tolerance relationship similar matrix to deal with the inconsistency problem of in the information table. And then obtains the core attributes of incomplete information systems via the tolerance relationship similar matrix. Finally, according to attribute frequency in the tolerance relationship similar matrix, as the heuristic knowledge, makes use of binsearch heuristic algorithm to calculate the candidate attribute expansion so that it can reduce the expansion times to speed up reduction. Experiment results show that the algorithm is simple and effective.