论文部分内容阅读
提出了一种基于粗糙集理论的数据挖掘模型，以利于信息不完备情况下的推理和决策问题的解决和研究。该模型从已知数据的初始决策系统出发，建立一系列的不同简化层次的子系统，然后推导出各个子系统的规则集，其中每条规则都有相应的置信度。在应用模型进行推理和决策分析时，用给定对象的信息与模型中相应节点的规则进行匹配，然后选用某种评判算法得出结论。给出了一个简单的例子来说明如何建立和应用这种数据挖掘模型。这样的模型可以很方便地根据给定的信息，在最符合的子系统上得出尽可能好的结论。
A data mining model based on rough set theory is proposed to facilitate the solution and research of inference and decision problems with incomplete information. The model starts from the original decision system of known data, establishes a series of subsystems with different simplified levels, and then deduces the rule sets of each subsystem, each of which has a corresponding degree of confidence. In the application of model for reasoning and decision analysis, the information of a given object is used to match the rules of the corresponding nodes in the model, and then the conclusion is obtained by using some kind of evaluation algorithm. A simple example is given to illustrate how to set up and apply this data mining model. Such a model makes it as easy as possible to draw the best possible conclusion on the most appropriate subsystems from the given information.