摘要
为使基于机器学习的兵棋演习战场态势分析理解取得更好结果,围绕兵棋演习数据特征提取问题,以深度学习方法为手段,提出了一种栈式稀疏降噪自编码网络模型,输入真实的兵棋演习数据进行了特征提取实验,通过分类精度表征了方法的效果,并进行了多种不同方法的对比实验,证明了深度学习方法的优势.
In order to improve the analysis and understanding of the wargame exercises battlefield situation based on machine learning,this paper focuses on the feature extraction problem of wargame exercises data, proposed a deep learning model of stacked sparse denoising autoencoder network, conducted the experiment with real wargame exercises data, demonstrated the effect by calculating the classification accuracy and prove its advantages by contrastive experiments based on different methods.
作者
郑书奎
吴琳
贺筱媛
ZHENG Shu-Kui WU Lin HE Xiao-Yuan(The Department of Information Operation Command Training, National Defense University, Beijing 100091, China)
出处
《指挥与控制学报》
2016年第3期194-201,共8页
Journal of Command and Control
基金
军民共用重大研究计划联合基金(U1435218)
国家自然科学基金(61403401)资助~~
关键词
深度学习
兵棋演习数据
特征提取
栈式稀疏降噪自编码网络
deep learning
wargame exercises data
feature extraction
stacked sparse denoising autoencoder network
