基于遗传算法和模糊粗糙集的知识约简

虽然粗糙集理论为处理离散属性提供了很好的工具,但它不能直接运用于具有连续变量的数据上面,而现实中的数据又包含着大量的连续变量.为了能够对连续属性集进行有效的知识约简,充分利用遗传算法的全局优化和并行计算的优点,结合模糊粗糙集的理论,对连续属性集进行知识约简,较粗糙集而言避开了连续属性的离散化过程,减少了信息损失,加快了约简速度,提高了决策支持度.首先利用一个仿真实例来验证该算法的有效性和快速性,然后把它运用于某一柴油机的故障数据集的约简,通过约简获得了影响输出故障模式的主要输入变量集,实现了数据的预处理,为进行柴油机的故障模式诊断提供了先决条件.

Knowledge-Reduction Based on GA and Fuzzy-rough Set

Given a dataset with discretized attribute values,by using the rough sets it is possible to find a subset(termed a reduct) of the original attributes using rough sets that are the most informative.All other attributes can be removed from the dataset with minimal information loss.However,it is often the case that the values of attributes may be both crisp and real-valued.In order to obtain better attribute reduction of the continuous dataset,genetic algorithm and fuzzy rough set were used.By making use of this method,the discretization process of continuous attributes was avoided,and the information loss was reduced,the reduction was quickened,the decision dependency was raised in comparison with the traditional rough set.Here a simulation example was used to test the efficiency of this method firstly.Then it was applied to the reduction of fault dataset about a diesel engine.The simulation result has shown that it can obtain those input features that are most predictive of a given outcome and realize the dataset preprocess,which is helpful to realize the fault diagnosis.