论文部分内容阅读
针对评估指标的重要性不一,且存在冗余问题,基于粗集可辨识矩阵,提出了一种计算指标属性重要度和约简的有效、简便算法,对样本信息进行约简,并计算约简后各指标的权重.其中,针对连续属性值离散化过程可能造成信息损失问题,采用了模糊C均值聚类算法离散化连续属性值.最后,建立了基于粗糙集和模糊C均值聚类的空战效能评估模型,并通过实例验证了该模型的可行性和有效性.
The importance of evaluating indicators is different, and there is redundancy problem. Based on the rough set distinguishable matrix, an efficient and simple algorithm for calculating the importance and reduction of index attributes is proposed to reduce the sample information and calculate the reduction The weight of each index afterwards.According to the possible loss of information caused by discretization of continuous attribute value, fuzzy C-means clustering algorithm is used to discretize continuous attribute values.Finally, an air combat based on rough set and fuzzy C-means clustering Effectiveness evaluation model, and verified the feasibility and effectiveness of the model through examples.