不协调决策信息系统的知识约简及决策规则优化研究

针对不协调决策信息系统的知识约简及决策规则的优化问题,引入分布约简和最大分布约简理论,提出一种基于分布区分对象集的知识约简算法,并得到具体的优化决策规则获取方法。该算法通过求解分布区分对象集和最小析取范式从而得到知识约简集,依据属性约简集挖掘出最优决策规则集。理论分析和实例结果表明该方法的有效性和实用性。

Research on Knowledge Reduction and Decision Rules Optimization In Inconsistent Decision Information Systems

Rule acquisition of decision information system is an important research content of data mining. According to the characteristics of the inconsistent decision information system, Introduced distribution reduction and maximum distribution reduction theory, obtain optimum attributes in system, and puts forward to a knowledge reduction algorithm based on distribution set of objects. The algorithm approach knowledge reduction sets by solving distribution set of objects and minimum disjunctive normal form, and mining optimal decision rules. Example analysis show that the method is effective and practical.