[1]于重重,吴子珺,谭励,等.非平衡集成迁移学习模型及其在桥梁结构健康监测中的应用[J].智能系统学报,2013,8(1):46-51.[doi:10.3969/j.issn.1673-4785.201208002]
YU Chongchong,WU Zijun,TAN Li,et al.Unbalanced integrated transfer learning model and its application to bridge structural health monitoring[J].CAAI Transactions on Intelligent Systems,2013,8(1):46-51.[doi:10.3969/j.issn.1673-4785.201208002]
点击复制
YU Chongchong,WU Zijun,TAN Li,et al.Unbalanced integrated transfer learning model and its application to bridge structural health monitoring[J].CAAI Transactions on Intelligent Systems,2013,8(1):46-51.[doi:10.3969/j.issn.1673-4785.201208002]
非平衡集成迁移学习模型及其在桥梁结构健康监测中的应用
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
8
期数:
2013年第1期
页码:
46-51
栏目:
学术论文—机器学习
出版日期:
2013-03-25
- Title:
- Unbalanced integrated transfer learning model and its application to bridge structural health monitoring
- 文章编号:
- 1673-4785(2013)01-0046-06
- 关键词:
- 非平衡集成迁移学习算法; SOM算法; 迁移学习模型; 桥梁结构健康监测
- Keywords:
- unbalanced integrated transfer learning algorithm; self organizing map algorithm; transfer learning model; bridge structural health monitoring
- 分类号:
- TP391
- 文献标志码:
- A
- 摘要:
- 在桥梁结构健康监测与状态评估过程中所获得的桥梁结构数据库常存在间断性异常或缺损，且不同样本分类数据不均匀，难以在信息缺失、数据分布失衡的情况下完成对桥梁结构健康的监测与状态评估.针对这一问题，在改进相似性度量函数的SOM聚类算法和非平衡集成迁移学习算法的基础上，提出了一种改进的迁移学习模型.通过对实际监测数据的分析，该迁移学习模型的分类精度随着目标数据集所占比例的不断增加而提高，验证了该模型的有效性和科学性.
- Abstract:
- The examination of bridge structural data obtained in the bridge structural health monitoring and condition assessment process had the problem of intermittent abnormalities or defects in the past. However, the classification of different samples of data is seen to be uneven, thus, making it difficult to complete structural health monitoring and condition assessment of the bridge under the condition of the absence of information and data distribution imbalance. In order to solve the problem mentioned above, this paper proposes an improved transfer learning model based on selforganizing map (SOM) clustering algorithm to improve the similarity measure function and unbalanced integration transfer learning algorithm. According to the analysis of actual monitoring data, the classification accuracy of the proposed transfer learning model increased as the increasing of the proportion of the target data set, validating the efficiency and scientificity of the proposed model.
- 参考文献/References:
-
［1］刘永前．大型桥梁结构健康监测技术研究与应用［D］．北京：北京交通大学, 2007: 1-12.
LIU Yongqian. Research and application on structural health monitoring of longspan bridge［D］. Beijing: Beijing Jiaotong University, 2007: 1-12.
［2］李鹏飞，吴太成．桥梁健康监测技术研究综述［J］．建筑监督检测与造价, 2010, 7(7): 28-31, 40.
LI Pengfei, WU Taicheng. A review of health monitoring techniques of bridge［J］. Supervision Test and Cost of Construction, 2010, 7(7): 28-31, 40.
［3］于重重，田蕊，谭励，等．非平衡样本分类的集成迁移学习算法［J］．电子学报, 2012, 40(7): 1358-1363.
YU Chongchong, TIAN Rui, TAN Li, et al. Integrated transfer learning algorithmic for unbalanced samples classification［J］. Acta Electronica Sinica, 2012, 40(7): 1358-1363.
［4］许至杰．迁移学习理论与算法研究［D］．上海：华东师范大学, 2012: 4-10.
XU Zhijie. Research about the theories and algorithms of transfer learning［D］. Shanghai: East China Normal University, 2012: 4-10.
［5］戴文渊．基于实例和特征的迁移学习算法研究［D］．上海：上海交通大学, 2008: 14-21.
DAI Wenyuan. Instancebased and featurebased transfer learning［D］. Shanghai: Shanghai Jiao Tong University, 2008: 14-21.
［6］UNKELBACH J, YI S, SCHMIDHUBER J. An EM based training algorithm for recurrent neural networks［J］. Proceedings of the 19th International Conference on Artificial Neural Networks. Berlin/Heidelberg: SpringerVerlag, 2009: 964-974.
［7］DAI Wenyuan, XUE Guirong, YANG Qiang, et al. Transferring naive Bayes classifiers for text classification［C］//Proceedings of the 22nd National Conference on Artificial Intelligence. Vancouver, Canada, 2007: 540-545.
［8］RAINA R, BATTLE A, LEE H, et al. Selftaught learning: transfer learning from unlabeled data［C］//The Twentyfourth International Conference on Machine Learning. Corvallis, USA, 2007: 759-766.
［9］刘伟，张化祥．数据集动态重构的集成迁移学习［J］．计算机工程与应用, 2010, 46(12): 126-128.
LIU Wei， ZHANG Huaxiang. Ensemble transfer learning algorithm based on dynamic dataset regroup［J］. Computer Engineering and Applications, 2010, 46(12): 126-128.
［10］邵昌昇，楼巍，严利民．高维数据中的相似性度量算法的改进［J］．计算机技术与发展, 2011, 2(2): 7-10.
SHAO Changsheng, LOU Wei, YAN Limin. Optimization of algorithm of similarity measurement in highdimensional data［J］. Computer Technology and Development, 2011, 2(2): 7-10.
［11］杜俊卫，李爱军．一种基于聚类的文本迁移学习算法［J］．计算机系统应用, 2010, 19(12): 238-241.
DU Junwei, LI Aijun. Transfer learning algorithm for text classification based on clustering［J］. Computer Systems & Applications, 2010, 19(12): 238-241.
［12］戴群，陈松灿，王喆．一个基于自组织特征映射网络的混合神经网络结构［J］．软件学报, 2009, 20(5): 1329-1336.
DAI Qun, CHEN Songcan, WANG Zhe. Hybrid neural network architecture based on selforganizing feature maps［J］. Journal of Software, 2009, 20(5): 1329-1336.
［13］王戒躁，钟继卫，王波．大跨桥梁健康监测系统设计构成及其进展［J］．桥梁建设, 2009(增刊2): 11-16.
WANG Jiezao, ZHONG Jiwei, WANG Bo. Designed components and development of health monitoring systems for long span bridges［J］. Bridge Construction, 2009(S2): 11-16.
［14］于重重，杨扬，涂序彦，等．DBSCAN算法在桥梁健康监测预测模型中的应用［J］．计算机工程与应用, 2008, 44(12): 224-227.
YU Chongchong, YANG Yang, TU Xuyan, et al. Application of DBSCAN algorithm in bridgehealth monitoring prediction model［J］. Computer Engineering and Applications, 2008, 44(12): 224-227.
备注/Memo
收稿日期： 2012-08-08.
网络出版日期：2013-01-25.
基金项目：国家自然科学基金资助项目（61070182）;北京市组织部优秀人才培养资助项目(2010D005003000008);北京市学科建设项目（PXM2012_014213_0000_74）.
通信作者：于重重.
E-mail: chongzhy@vip.sina.com.
作者简介：于重重，女，1971年生，教授，硕士生导师，主要研究方向为智能信息处理与模式识别、复杂实时监测系统预测与评估等.参与“十一五”重大科技支撑项目子课题、国家自然科学基金项目多项，获部科技进步奖1项.发表学术论文40余篇，其中被EI检索20余篇，合作出版著作4部.
吴子珺，男，1989年生，硕士研究生，主要研究方向为机器学习、人工智能等.
谭励，女，1980年生，副教授，博士，主要研究方向为智能信息处理、机器学习、计算机视觉、无线传感器网络等.作为项目负责人和主研人员参与国家自然科学基金项目1项、北京市优秀人才项目1项.发表学术论文20余篇，出版著作1部.
涂序彦，男，1935年生，教授，博士生导师，北京科技大学计算机与系统科学研究所所长，中国人工智能学会荣誉理事长，北京科技大学信息工程学院特聘教授.创立多变量协调控制理论、最经济控制理论;倡导人体控制论、大系统控制论新学科, 开拓人工智能、专家系统、智能控制、智能管理、人工生命新方法、新技术.荣获国家科技攻关重大成果奖，冶金部、机械部、电子部等科技进步奖多项.发表学术论文300余篇，出版专著12部.
田蕊，女，1986年生，硕士研究生，主要研究方向为机器学习、人工智能等.
更新日期/Last Update:
2013-04-12