摘要
Relief是公认的效果较好的filter式特征评估方法,但存在特征权值随样本波动的问题,导致识别准确率的下降。提出了一种基于均值-方差模型的特征权值优化算法,采用样本区分能力的平均贡献值的期望和组合贡献值的波动作为特征评估的依据,使得特征选择的结果更加稳定与准确。基于实地采集的地面运动目标的震动信号进行特征选择与分类学习实验,结果表明,该算法得到的特征子集比Relief具有更好的目标区分能力。
Relief is a feature evaluation method which performs well, while the weight of feature could fluctuate with the samples, which lead to poor recognition accuracy. To solve this problem, a novel feature selection algorithm based on Mean-Variance model is presented. It takes both the mean and the variance into account as the criterion of feature evaluation, which makes the result more stable and accurate. Based on real seismic signals of ground targets, experiment results indicate that the subsets of feature generated by proposed algorithm have better performance.
出处
《系统仿真技术》
2013年第3期224-228,共5页
System Simulation Technology
基金
国家科技重大专项资助项目(2010ZX03006-004)
国家重点基础研究发展计划资助项目(2011CB302906)
