Anomaly Detection Method Based on One-Class Random Forest with Applications

doi:10.7652/xjtuxb202002001

您当前的位置：

首页 >

文章列表页 >

Anomaly Detection Method Based on One-Class Random Forest with Applications

更新时间：2025-05-08

- Anomaly Detection Method Based on One-Class Random Forest with Applications
- Vol. 54, Issue 2, Pages: 1-8+157(2020)
- 作者机构：
  
  西安交通大学机械制造系统工程国家重点实验室,西安,710049
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb202002001
  CLC： TH17
- Online First：10 February 2020，
  
  Published：2020
- 稿件说明：
移动端阅览
Anomaly Detection Method Based on One-Class Random Forest with Applications[J]. 2020, 54(2): 1-8+157.
DOI：

Anomaly Detection Method Based on One-Class Random Forest with Applications[J]. 2020, 54(2): 1-8+157. DOI： 10.7652/xjtuxb202002001.

摘要

针对随机森林算法不能处理异常检测问题的局限

提出了一种基于改进格雷厄姆扫描法的单类随机森林

实现了随机森林在只有单类样本时的分类应用。在格雷厄姆扫描法的算法原理以及操作流程的基础上

引入了边界软化率的概念

增大了数据点外边界的柔性。利用射线法生成与输入样本反分布的数据集

使得传统的随机森林模型经过训练后成为拥有精细决策边界的单类随机森林。训练好的模型输出待测数据的异常概率。在XJTU-SY轴承数据集上验证了所提方法对于滚动轴承状态监测的有效性。分析结果表明

单类随机森林能准确分离正常运行数据与退化数据

并且可以通过调整边界软化率来实现异常值检测准确率与召回率的平衡。当预测概率阈值取0.5时

硬边界的单类随机森林可以实现98.37%的检测准确率以及0.972的召回率

而0.05的边界软化率可以得到最小的退化后期预测均方根误差

较硬边界降低1.01%。随着阈值的增大

边界软化率提供了有力的召回率保障。

Abstract

Aiming at the limitation that the random forest algorithm cannot deal with the problem of anomaly detection

a one-class random forest based on improved Graham scanning method is proposed

which reali-es the classification application of the random forest with only a single class of samples. Following the principle and process of Graham scanning method

the concept of boundary softening ratio is introduced to increase the flexibility of the outer boundary of data points. Ray casting is used to generate data set with inverse distribution of input samples to make the traditional random forest model become a one-class random forest with fine decision boundary after training

which outputs the abnormal probability of the data to be tested. The effectiveness of this method for the condition monitoring of rolling bearings is verified on the XJTU-SY bearing data set. The results show that the one-class random forest can accurately separate normal data from degradation data. Adjusting the boun

关键词

Keywords

references

邱卫, 杨英杰, 汪永伟, 等. 基于改进遗传算法和隐Markov模型的协议异常检测方法 [J]. 计算机应用研究, 2016, 33(4): 1164-1168.

QIU Wei, YANG Yingjie, WANG Yongwei, et al. Protocol anomaly detection method based on improved genetic algorithms and hidden Markov model [J]. Application Research of Computers, 2016, 33(4): 1164-1168.

曹玉苹, 卢霄, 田学民, 等. 基于动态单类随机森林的非线性过程监控方法 [J]. 化工学报, 2017, 68(4): 1459-1465.

CAO Yuping, LU Xiao, TIAN Xuemin, et al. Nonlinear process monitoring using dynamic one-class random forest [J]. CIESC Journal, 2017, 68(4): 1459-1465.

史忠奎, 李培军, 罗伦, 等. 基于形态学属性剖面和单类随机森林分类的道路路域新增建筑物提取方法 [J]. 北京大学学报(自然科学版), 2018, 54(1): 105-114.

SHI Zhongkui, LI Peijun, LUO Lun, et al. A method for extraction of newly-built buildings in road region using morphological attribute profiles and one-class random forest [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2018, 54(1): 105-114.

周绪川, 蔡利平. 基于局部异常因子的无监督学习集成异常检测 [J]. 西南民族大学学报(自然科学版), 2012, 38(3): 444-448.

ZHOU Xuchuan, CAI Liping. Ensemble anomaly detection with unsupervised learning method based on local outlier factor algorithm [J]. Journal of Southwest University for Nationalities(Natural Science Edition), 2012, 38(3): 444-448.

LIU F T, TING K M, ZHOU Z H. Isolation forest [C]∥2008 Eighth IEEE International Conference on Data Mining. Piscataway, NJ, USA: IEEE, 2008: 413-422.

尚文利, 李琳, 万明, 等. 基于优化单类支持向量机的工业控制系统入侵检测算法 [J]. 信息与控制, 2015, 44(6): 678-684.

SHANG Wenli, LI Lin, WAN Ming, et al. Intrusion detection algorithm based on optimized one-class support vector machine for industrial control system [J]. Information and Control, 2015, 44(6): 678-684.

DéSIR C, BERNARD S, PETITJEAN C, et al. One class random forests [J]. Pattern Recognition, 2013, 46(12): 3490-3506.

胡昌华, 施权, 司小胜, 等. 数据驱动的寿命预测和健康管理技术研究进展 [J]. 信息与控制, 2017, 46(1): 72-82.

HU Changhua, SHI Quan, SI Xiaosheng, et al. Data-driven life prediction and health management: state of the art [J]. Information and Control, 2017, 46(1): 72-82.

张小丽, 王保建, 马猛, 等. 滚动轴承寿命预测综述 [J]. 机械设计与制造, 2015(10): 221-224.

ZHANG Xiaoli, WANG Baojian, MA Meng, et al. A review of life prediction for roller bearing [J]. Machinery Design Manufacture, 2015(10): 221-224.

BREIMAN L. Random forests [J]. Machine Learning, 2001, 45(1): 5-32.

李航. 统计学习方法 [M]. 北京: 清华大学出版社, 2012: 67-71.

毛鹏. 快速凸包计算实现及其应用 [D]. 西安: 西安电子科技大学, 2013: 15-20.

向俊, 王静, 夏幼明. 判断点与多边形拓扑关系的改进算法 [J]. 计算机工程与设计, 2014, 35(5): 1732-1737.

XIANG Jun, WANG Jing, XIA Youming. Improved algorithm to determine topological relationship between point and polygon [J]. Computer Engineering and Design, 2014, 35(5): 1732-1737.

周志华. 机器学习 [M]. 北京: 清华大学出版社, 2016: 225-240.

WANG B, LEI Y G, LI N P, et al. A hybrid prognostics approach for estimating remaining useful life of rolling element bearings [J]. IEEE Transactions on Reliability, 2018: 1-12.

张西宁, 雷威, 李兵. 主分量分析和隐马尔科夫模型结合的轴承监测诊断方法 [J]. 西安交通大学学报, 2017, 51(6): 1-7, 109.

ZHANG Xining, LEI Wei, LI Bing. Bearing fault detection and diagnosis method based on principal component analysis and hidden Markov model [J]. Journal of Xi’an Jiaotong University, 2017, 51(6): 1-7, 109.

CHUAN L, SANCHEZ R V, ZURITA G, et al. Gearbox fault diagnosis based on deep random forest fusion of acoustic and vibratory signals [J]. Mechanical Systems and Signal Processing, 2016, 76/77: 283-293.

胡淼, 王开军, 李海超, 等. 模糊树节点的随机森林与异常点检测 [J]. 南京大学学报(自然科学), 2018, 54(6): 1141-1151.

HU Miao, WANG Kaijun, LI Haichao, et al. A random forest algorithm based on fuzzy tree node for anomaly detection [J]. Journal of Nanjing University(Natural Science), 2018, 54(6): 1141-1151.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Image Super Resolution Algorithm Based on Receptive Field Optimization and Progressive Feature Fusion

Research on Railway Track Foreign Object Intrusion Detection Based on Multi-Scale Feature Fusion

A Method of Image Content Matching with Lattice Closeness and Multi-Feature Fusion

Spatial Attention Mechanism with Global Characteristics

A Structure-Semantics-Residual Collaborative Image Compression Framework：Towards Privacy Protection and High-Fidelity Reconstruction

Related Author

WU Hongwu

GAI Shaoyan

DA Feipeng

WANG Nan

HOU Tao

NIU Hongxia

肖满生

肖哲

Related Institution

School of Automation, Southeast University

Key Laboratory of Measurement and Control of Complex Systems of Engineering, Ministry of Education, Southeast University

(School of Automation and Electrical Engineering, Lanzhou Jiaotong University,,)

School of Computer Science, Hunan University of Technology

School of Information and Communications Engineering, Xi’an Jiaotong University

⁰