摘要:为进一步提高变压器故障诊断准确率,提出一种基于ReliefF-mRMR 与 IAO-SVM 结合的变压器故障诊断模型。采用ReliefF和最大相关最小冗余 (mRMR) 算法对变压器故障数据进行特征优选;引入混沌反向学习和自适应混合变异策略改进天鹰优化算法,并对最优...
在所述的步骤四中,构造训练样本数据集o、特征参数集f、目标类别集c,定义加权的人体生理特征参数集,初始的为空,将训练样本数据集o、特征参数集f、目标类别集c输入到改进后的rrelieff算法,并由以下的计算公式得到每一个特征参数的权重值,计算公式为:,表示在不同的预测值条件下的权重,表示在不同的特征条件下的权重...
Alzaqebah, "Hybridizing relieff, mRMR filters and GA wrapper approaches for gene selection," J. Theor. Appl. Inf. Technol., vol. 46, no. 2, pp. 1034-1039, 2012.Shreem S, Abdullah S, Nazri M, Alzaqebah M, Hybridizing Relief F. MRMR filters and GA wrapper approaches for gene ...
1.一种基于改进RReliefF和mRMR相结合的人体特征参数选择方法,其特征是包括以下步骤: 步骤一:计算样本的欧氏距离和形态距离; 步骤二:标准化并归一化欧氏距离和形态距离; 步骤三:构建样本相似距离模型来改进RReliefF算法; 步骤四:利用改进后的RReliefF算法来计算各个特征权重,去除不相关特征; 步骤五:利用mRMR算法来计...
ReliefFMinimum redundancy maximum relevanceFeature selectionCognition may be defined as a set of mental activities or processes which deals with knowledge, attention, memory and working memory, reasoning and computation, and judgement and evaluation. In this paper, we aim to study two distinctive ...