Some Technologies And Application Research Of Rough Set Theory

Rough Set (RS) theory- was put forward by Zdzislaw Pawlak in 1982. After about 20 years of developing, it has received fruitful achievements in both of theories and applications. RS doesn't depend on additional information beyond the data set, which is a potent tool for dealing with vague, imprecise, incomplete and uncertain data, and is a new branch in .intelligent information processing research.In this dissertation, the reducts of information system, clustering analysis and integration rough set with fuzzy neural network, multilayer classifier, respective etc. are researched deeply, and put forward some new approaches and apply them to practical fields. Firstly, this dissertation introduces the scheme of the rough set theory and its application. Then, several improved search methods of reduce of information systems are put forward. And then some new methods are given to deal with the classification and discretization of continuous attributes. In addition, two new schemes of knowledge encoding by integrating the neural networks with rough set theory are described. Finally, an application to cost analysis of product using rough set theory is given.The major innovations in the present study are as follows:1 )Put forward the improved search method of reduct of information systems, such as pretreatment strategy of discernibility matrices by using lower orders and sparsity matrices, heuristic search algorithm of minimal reduct based on attribute frequency heuristic function.2)Based on k-means algorithm, a clustering analysis method using genetic algorithm is studied. Experiment results proved its validity and the method can be applied to solve the clustering problem.3)Based on the uncertain measurement of information, according to the feature of datadistributing in database with continuous attributes and a criterion function