• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Computer Vision, Image and Deep Learning >Research on Application of Weighted Sampling Bayesian Network Parameter Learning in Bridge Condition Assessment
【24h】

Research on Application of Weighted Sampling Bayesian Network Parameter Learning in Bridge Condition Assessment

机译：加权抽样贝叶斯网络参数学习在桥梁条件评估中的应用研究

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

EM algorithm is a traditional Bayesian network parameter learning method. In practical applications, due to the complex structure of the network and large data deviations, the computational complexity problem is the major factor limiting its development. Aiming at the efficiency of EM algorithm, this paper proposes an improved algorithm of sampling EM with measurement index completion. The algorithm first modifies the iterative judgment of the EM algorithm, and proposes a judgment method for setting a measurement index for each missing sample to improve the calculation efficiency; meanwhile, Gibbs sampling is introduced to optimize the sample completion calculation, so as to reduce the overall computational complexity of the algorithm. Finally, the algorithm is used to evaluate the health status of Wenhui Bridge in Liuzhou to prove the practicability of the algorithm.
机译：EM算法是一种传统的贝叶斯网络参数学习方法。在实际应用中，由于网络结构复杂和大数据偏差，计算复杂性问题是限制其发展的主要因素。旨在EM算法的效率，本文提出了一种利用测量指数完成的采样算法。该算法首先修改EM算法的迭代判断，并提出了一种判断方法，用于为每个缺失样本设置测量指数以提高计算效率;同时，引入了GIBBS采样以优化采样完成计算，从而降低算法的整体计算复杂性。最后，该算法用于评估柳州文汇桥的健康状况证明算法的实用性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号