1. 军械工程学院四系,河北,石家庄,050003
2. 电子信息系统复杂电磁环境效应国家重点实验室,河南,洛阳,471003
3. 军械工程学院四系,河北,石家庄,050003
4. 电子信息系统复杂电磁环境效应国家重点实验室,河南,洛阳,471003
网络出版:2014-02-25,
纸质出版:2014
移动端阅览
李兵, 董俊, 刘鹏远, 等. 模糊格构造型形态神经网络[J]. 电子学报, 2014,42(2):319-327.
LI Bing, DONG Jun, LIU Peng-yuan, et al. Fuzzy Lattice Constructive Morphological Neural Network[J]. Acta Electronica Sinica, 2014, 42(2): 319-327.
针对构造型形态神经网络(CMNN)决策函数的局限性,提出了一种模糊格构造型形态神经网络(FL-CMNN);该模型在利用训练好的CMNN进行分类时,引入模糊格包容性测度计算测试样本属于各超盒的隶属度值.采用仿真数据集对提出的FL-CMNN模型进行了评价,并与原始的CMNN和传统的人工神经网络、支持向量机、最近邻分类器进行了对比;试验结果表明,FL-CMNN在测试精度上明显优于原始的CMNN,训练时间远远低于传统的神经网络和支持向量机,而分类精度丝毫不亚于传统的神经网络和支持向量机.
A novel neural network model named fuzzy lattice constructive morphological neural network (FL-CMNN) is presented to overcome the deficiency of the original constructive morphological neural network(CMNN)
which suffers for the problem of decision function in classification phase.The fuzzy lattice inclusion measure function is introduced to calculate the membership of testing sample belong to the hyper-boxes trained by the CMNN.Three standard datasets are employed to evaluate and compare the presented FL-CMNN with the CMNN
artificial neural network(ANN)
support vector machine(SVM)and K nearest neighbor(KNN)classifiers.Experimental results have revealed that the presented FL-CMNN yields better performance than the original CMNN model.It also achieved comparative classification accuracies with much lower computational cost than traditional ANN and SVM model.
0
浏览量
1303
下载量
2
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621