【摘要】 提出了一种基于决策属性支持度的属性相对约简算法。通过引入决策属性支持度对不完备决策表中属性的重要性进行了定义,并以此作为启发信息进行属性的选择,该算法的时间复杂度是多项式的。寻找决策表中最小相对约简问题是典型的NP-hard问题,采用该算法可降低问题复杂度。通过实例说明,该算法能得到不完备决策表的最小相对约简。更多还原

【Abstract】 A kind of attribute relative reduction for decision attribute support degree was proposed.States the significance of attributes in the incomplete decision table by introducing decision attribute support degree,which is then used as heuristic information for selecting attributes,this algorithm has a polynomial time complexity.Seeking the minimal relative reduction in a decision table is typically a NP-hard problem,and its complexity can be reduced by using the algorithm provided in this paper.An example shows this algorithm can achieve the minimal relative reduction of incomplete decision table.更多还原