- Full text PDF available (0)
For inconsistent decision tables, this paper presents a new data reduction idea. Compared with the traditional data reduction method, the first step in this paper is attribute value reduction, then this paper simply deletes the inconsistent individuals sets. The last step is attribute reduction, which could avoid the negative factor of inconsistent… (More)
The granulating is the segmentation from the whole into the parts. The elementary granulation and the granulation are defined in this paper, and with construction of the granular network, the relevant knowledge of graph theory is applied to granular network and its computing. In the paper it is shown using the example that it is feasible and effective that… (More)
This paper defines some basic concepts. Then, this paper introduces how to convert any decision table into certain granular graph. We give visual presentation of decision table. Some relevant knowledge of graph theory is applied to granular graph and its computing for the inconsistent decision table. It is feasible and effective when granular graph is… (More)
This paper proposes a new reduction, that is, considers the relation of attribute reduction and attribute value reduction. When making reduction, both the attribute reduction and the attribute value reduction can be done simultaneously in the process of reduction in a decision table. In comparison with the traditional methods, the complicated comparison… (More)
The paper mainly supplements the definition of Granular sets which was proposed by predecessors. If let X be given, X is the subset of one or more granules and each granule is not the subset of X on the tolerance relation of Gr, we define some concepts of lower and upper approximation of X, and discuss its properties.