2.765

2022影响因子

(CJCR)

  • 中文核心
  • EI
  • 中国科技核心
  • Scopus
  • CSCD
  • 英国科学文摘

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

面向入侵检测的元图神经网络构建与分析

王振东 徐振宇 李大海 王俊岭

王振东, 徐振宇, 李大海, 王俊岭. 面向入侵检测的元图神经网络构建与分析. 自动化学报, 2023, 49(7): 1530−1548 doi: 10.16383/j.aas.c200819
引用本文: 王振东, 徐振宇, 李大海, 王俊岭. 面向入侵检测的元图神经网络构建与分析. 自动化学报, 2023, 49(7): 1530−1548 doi: 10.16383/j.aas.c200819
Wang Zhen-Dong, Xu Zhen-Yu, Li Da-Hai, Wang Jun-Ling. Construction and analysis of meta graph neural network for intrusion detection. Acta Automatica Sinica, 2023, 49(7): 1530−1548 doi: 10.16383/j.aas.c200819
Citation: Wang Zhen-Dong, Xu Zhen-Yu, Li Da-Hai, Wang Jun-Ling. Construction and analysis of meta graph neural network for intrusion detection. Acta Automatica Sinica, 2023, 49(7): 1530−1548 doi: 10.16383/j.aas.c200819

面向入侵检测的元图神经网络构建与分析

doi: 10.16383/j.aas.c200819
基金项目: 国家自然科学基金(62062037, 61763017), 江西省自然科学基金(20212BAB202014, 20181BBE58018)资助
详细信息
    作者简介:

    王振东:博士, 江西理工大学信息工程学院副教授. 主要研究方向为无线传感器网络, 智慧物联网, 认知计算, 大数据与信息安全. 本文通信作者. E-mail: wangzhendong@hrbeu.edu.cn

    徐振宇:江西理工大学信息工程学院硕士研究生. 主要研究方向为信息安全. E-mail: xuzhenyu0208@163.com

    李大海:博士, 江西理工大学信息工程学院副教授. 主要研究方向为分布式系统服务质量(QoS)控制, 分布式系统自学习资源调度控制. E-mail: dlai6535@aliyun.com

    王俊岭:博士, 江西理工大学信息工程学院副教授. 主要研究方向为分布式计算, 容错, 计算机视觉. E-mail: wangjunling@jxust.edu.cn

Construction and Analysis of Meta Graph Neural Network for Intrusion Detection

Funds: Supported by National Natural Science Foundation of China (62062037, 61763017) and Natural Science Grant of Jiangxi Province (20212BAB202014, 20181BBE58018)
More Information
    Author Bio:

    WANG Zhen-Dong Ph.D., associate professor at the School of Information Engineering, Jiangxi University of Science and Technology. His research interest covers wireless sensor networks, smart internet of things, cognitive computing, big data, and information security. Corresponding author of this paper

    XU Zhen-Yu Master student at the School of Information Engineering, Jiangxi University of Science and Technology. His main research interest is information security

    LI Da-Hai Ph.D., associate professor at the School of Information Engineering, Jiangxi University of Science and Technology. His research interest covers distributed system quality of service (QoS) control, and distributed system self-learning resource scheduling control

    WANG Jun-Ling Ph.D., associate professor at the School of Information Engineering, Jiangxi University of Science and Technology. His research interest covers distributed computing, fault tolerance, and computer vision

  • 摘要: 网络入侵样本数据特征间存在未知的非欧氏空间图结构关系, 深入挖掘并利用该关系可有效提升网络入侵检测方法的检测效能. 对此, 设计一种元图神经网络(Meta graph neural network, MGNN), MGNN能够对样本数据特征内部隐藏的图结构关系进行挖掘与利用, 在应对入侵检测问题时优势明显. 首先, 设计元图网络层(Meta graph network layer, MGNL), 挖掘出样本数据特征内部隐藏的图结构关系, 并利用该关系对样本数据的原始特征进行更新; 然后, 针对MGNN存在的图信息传播过程中父代信息湮灭现象提出反信息湮灭策略, 并设计了注意力损失函数, 简化MGNN中实现注意力机制的运算过程. KDD-NSL、UNSW-NB15、CICDoS2019数据集上的实验表明, 与经典深度学习算法深度神经网络 (Deep neural network, DNN)、卷积神经网络(Convolutional neural network, CNN)、循环神经网络(Recurrent neural network, RNN)、长短期记忆(Long short-term memory, LSTM)和传统机器学习算法支持向量机(Support vector machine, SVM)、决策树(Decision tree, DT)、随机森林(Random forest, RF)、K-最近邻(K-nearest neighbor, KNN)、逻辑回归(Logistic regression, LR)相比, MGNN在准确率、F1值、精确率、召回率评价指标上均具有良好效果.
  • 图  1  MGNN结构与处理流程

    Fig.  1  Structure and processing flow of MGNN

    图  2  MGNL中单代父子结点间信息传递结构

    Fig.  2  Information transfer process of parent-child node between single generation in MGNL

    图  3  MGNN祖孙结点间信息传递结构

    Fig.  3  Information transfer process between grandparents and grandchildren in MGNN

    图  4  MGNN运行流程图

    Fig.  4  Operation flow chart of MGNN

    图  5  MGNNSB描述

    Fig.  5  Description of MGNNSB structure

    图  6  各神经网络对UNSW_NB15进行二分类

    Fig.  6  Each neural network performs a binary classification experiment on the UNSW_NB15

    图  7  各神经网络对NSL_KDD进行二分类

    Fig.  7  Each neural network performs a binary classification experiment on the NSL_KDD

    图  8  各神经网络对UNSW_NB15进行多分类

    Fig.  8  Each neural network performs multi-classification experiments on the UNSW_NB15

    图  9  各神经网络对NSL_KDD进行多分类

    Fig.  9  Each neural network performs multi-classification experiments on the NSL_KDD

    图  10  MGNN与最新入侵检测算法对比

    Fig.  10  Performance comparison between MGNN and the latest intrusion detection algorithms on different datasets

    图  11  各神经网络对UNSW_NB15进行二分类

    Fig.  11  Each neural network performs a binary classification experiment on the UNSW_NB15

    图  12  各神经网络对NSL_KDD进行二分类

    Fig.  12  Each neural network performs a binary classification experiment on the NSL_KDD

    图  13  各神经网络对UNSW_NB15进行多分类

    Fig.  13  Each neural network performs multi-classification experiments on the UNSW_NB15

    图  14  各神经网络对NSL_KDD进行多分类

    Fig.  14  Each neural network performs multi-classification experiments on the NSL_KDD

    图  15  注意力损失函数对MGNN的影响

    Fig.  15  The effect of attention loss function on MGNN

    表  1  MGNN1 ~ MGNN9网络各参数设置

    Table  1  Various parameter settings in the MGNN1 ~ MGNN9 networks

    网络类别MGNNSBNnPnUnits$\alpha $Activation参数量
    MGNN11421641tanh287509
    26411281tanh
    312812681tanh
    426812681tanh
    MGNN31423641tanh287509
    26431281tanh
    312832681tanh
    426832681tanh
    MGNN51425641tanh287509
    26451281tanh
    312852681tanh
    426852681tanh
    MGNN71427641tanh287509
    26471281tanh
    312872681tanh
    426872681tanh
    MGNN91429641tanh287509
    26491281tanh
    312892681tanh
    426892681tanh
    下载: 导出CSV

    表  2  各算法对UNSW_NB15数据集二分类测试的结果

    Table  2  The experimental results of the binary classification test of each algorithm on the UNSW_NB15 dataset

    算法AccuracyPrecisionRecallF1-score
    MGNN10.9020.9100.9120.911
    MGNN30.9290.9470.9240.935
    MGNN50.9400.9590.9310.945
    MGNN70.9430.9610.9330.947
    MGNN90.9450.9640.9350.949
    DNN0.8900.9010.8980.900
    CNN0.8530.8980.8270.861
    RNN0.7090.7220.7660.744
    LSTM0.8130.8770.7680.819
    RF0.9030.9880.8670.924
    LR0.7430.9550.6530.775
    KNN0.8100.9320.7780.848
    DT0.8970.9820.8640.919
    SVM_RBF0.6530.9980.4920.659
    下载: 导出CSV

    表  3  各算法对UNSW_NB15数据集多分类测试的结果

    Table  3  The experimental results of the multi-classification test of each algorithm on the UNSW_NB15 dataset

    算法AccuracyPrecisionRecallF1-score
    MGNN10.7720.7350.7720.743
    MGNN30.8160.7870.8160.797
    MGNN50.8260.8010.8260.812
    MGNN70.8400.8240.8400.829
    MGNN90.8360.8150.8360.824
    DNN0.7620.7180.7620.724
    CNN0.6160.5300.6160.501
    RNN0.6400.4430.6400.521
    LSTM0.6600.5610.6600.566
    RF0.7550.7550.7550.724
    LR0.5380.4140.5380.397
    KNN0.6220.5780.6220.576
    DT0.7330.7210.7330.705
    SVM_RBF0.5810.5860.5810.496
    下载: 导出CSV

    表  4  各算法对NSL_KDD数据集二分类测试的结果

    Table  4  The experimental results of the binary classification test of each algorithm on the NSL_KDD dataset

    算法AccuracyPrecisionRecallF1-score
    MGNN10.9850.9850.9820.984
    MGNN30.9860.9890.9810.985
    MGNN50.9860.9880.9810.985
    MGNN70.9900.9950.9850.990
    MGNN90.9720.9710.9700.970
    DNN0.9790.9750.9800.978
    CNN0.9790.9880.9670.977
    RNN0.9270.9250.9190.922
    LSTM0.9100.8950.9150.905
    RF0.9290.9460.9190.933
    LR0.8260.9150.7440.820
    KNN0.9100.9260.9050.915
    DT0.9300.9280.9430.935
    SVM_RBF0.8370.7690.9930.867
    下载: 导出CSV

    表  5  各算法对NSL_KDD数据集多分类测试的结果

    Table  5  The experimental results of the multi-classification test of each algorithm on the NSL_KDD dataset

    算法AccuracyPrecisionRecallF1-score
    MGNN10.9860.9850.9860.985
    MGNN30.9870.9870.9870.987
    MGNN50.9860.9850.9860.985
    MGNN70.9750.9670.9750.971
    MGNN90.5330.2840.5330.371
    DNN0.9570.9550.9570.955
    CNN0.9700.9690.9700.968
    RNN0.8930.8840.8930.887
    LSTM0.8650.8660.8650.838
    RF0.7530.8140.7530.715
    LR0.6120.5090.6120.530
    KNN0.7310.7200.7310.684
    DT0.7630.7670.7630.728
    SVM_RBF0.7020.6890.7020.656
    下载: 导出CSV

    表  6  各算法对CICDoS2019数据集测试

    Table  6  Test results of each algorithm on the CICDoS2019 dataset

    算法AccuracyPrecision Recall F1-score
    AttackBenignAttackBenignAttackBenign
    MGNN120.870.991.00 0.790.93 0.880.96
    NB0.571.000.530.171.000.290.69
    DT0.770.700.980.990.540.820.70
    LR0.950.930.990.990.910.960.95
    RF0.861.000.780.741.000.850.88
    Booster0.840.760.990.990.670.860.80
    SVM0.930.990.880.880.990.930.93
    DDoSNet0.990.991.000.990.990.990.99
    下载: 导出CSV

    表  7  MGNN9、MGNN9_alpha网络对UNSW_NB15数据集二分类测试的结果

    Table  7  MGNN9, MGNN9_alpha networks on the UNSW_NB15 dataset binary classification test results

    算法AccuracyPrecisionRecallF1-score
    MGNN90.9450.9640.9350.949
    MGNN9_alpha0.9510.9720.9390.955
    下载: 导出CSV

    表  8  MGNN9、MGNN9_alpha网络对UNSW_NB15数据集多分类测试的结果

    Table  8  MGNN9, MGNN9_alpha networks on the UNSW_NB15 dataset multi-classification test results

    算法AccuracyPrecisionRecallF1-score
    MGNN90.8360.8150.8360.824
    MGNN9_alpha0.8460.8310.8460.837
    下载: 导出CSV

    表  9  MGNN9、MGNN9_alpha网络对NSL_KDD数据集二分类测试的结果

    Table  9  MGNN9, MGNN9_alpha networks on the NSL_KDD dataset binary classification test results

    算法AccuracyPrecisionRecallF1-score
    MGNN90.9720.9710.9700.970
    MGNN9_alpha0.9920.9930.9900.991
    下载: 导出CSV

    表  10  MGNN9、MGNN9_alpha网络对NSL_KDD数据集多分类测试的结果

    Table  10  MGNN9, MGNN9_alpha networks on the NSL_KDD dataset multi-classification test results

    算法AccuracyPrecisionRecallF1-score
    MGNN90.5330.2840.5330.371
    MGNN9_alpha0.9870.9870.9870.986
    下载: 导出CSV
  • [1] Tsai C F, Hsu Y F, Lin C Y, Lin W Y. Intrusion detection by machine learning: A review. Expert Systems With Applications, 2009, 36(10): 11994-12000 doi: 10.1016/j.eswa.2009.05.029
    [2] 任家东, 刘新倩, 王倩, 何海涛, 赵小林. 基于KNN离群点检测和随机森林的多层入侵检测方法. 计算机研究与发展, 2019, 56(3): 566-575

    Ren Jia-Dong, Liu Xin-Qian, Wang Qian, He Hai-Tao, Zhao Xiao-Lin. An multi-level intrusion detection method based on KNN outlier detection and random forests. Journal of Computer Research and Development, 2019, 56(3): 566-575
    [3] Ahmad I, Basheri M, Iqbal M J, Rahim A. Performance comparison of support vector machine, random forest, and extreme learning machine for intrusion detection. IEEE Access, 2018, 6: 33789-33795 doi: 10.1109/ACCESS.2018.2841987
    [4] Mabu S, Gotoh S, Obayashi M, Kuremoto T. A random-forests-based classifier using class association rules and its application to an intrusion detection system. Artificial Life and Robotics, 2016, 21(3): 371-377 doi: 10.1007/s10015-016-0281-x
    [5] 缪祥华, 单小撤. 基于密集连接卷积神经网络的入侵检测技术研究. 电子与信息学报, 2020, 42(11): 2706-2712

    Miao Xiang-Hua, Shan Xiao-Che. Research on intrusion detection technology based on densely connected convolutional neural networks. Journal of Electronics & Information Technology, 2020, 42(11): 2706-2712
    [6] 王振东, 刘尧迪, 杨书新, 王俊岭, 李大海. 基于天牛群优化与改进正则化极限学习机的网络入侵检测. 自动化学报, 2022, 48(12): 3024-3041

    Wang Zhen-Dong, Liu Yao-Di, Yang Shu-Xin, Wang Jun-Ling, Li Da-Hai. Network intrusion detection based BSO and improved RELM. Acta Automatica Sinica, 2022, 48(12): 3024-3041
    [7] 张颐康, 张恒, 刘永革, 刘成林. 基于跨模态深度度量学习的甲骨文字识别. 自动化学报, 2021, 47(4): 791-800

    Zhang Yi-Kang, Zhang Heng, Liu Yong-Ge, Liu Cheng-Lin. Oracle character recognition based on cross-modal deep metric learning. Acta Automatica Sinica, 2021, 47(4): 791-800
    [8] 徐鹏斌, 瞿安国, 王坤峰, 李大字. 全景分割研究综述. 自动化学报, 2021, 47(3): 549-568

    Xu Peng-Bin, Qu An-Guo, Wang Kun-Feng, Li Da-Zi. A survey of panoptic segmentation methods. Acta Automatica Sinica, 2021, 47(3): 549-568
    [9] 徐聪, 李擎, 张德政, 陈鹏, 崔家瑞. 文本生成领域的深度强化学习研究进展. 工程科学学报, 2020, 42(4): 399-411

    Xu Cong, Li Qing, Zhang De-Zheng, Chen Peng, Cui Jia-Rui. Research progress of deep reinforcement learning applied to text generation. Chinese Journal of Engineering, 2020, 42(4): 399-411
    [10] 宋勇, 侯冰楠, 蔡志平. 基于深度学习特征提取的网络入侵检测方法. 华中科技大学学报(自然科学版), 2021, 49(2): 115-120

    Song Yong, Hou Bing-Nan, Cai Zhi-Ping. Network intrusion detection method based on deep learning feature extraction. Journal of Huazhong University of Science and Technology (Natural Science Edition), 2021, 49(2): 115-120
    [11] Gao L G, Chen P Y, Yu S M. Demonstration of convolution kernel operation on resistive cross-point array. IEEE Electron Device Letters, 2016, 37(7): 870-873 doi: 10.1109/LED.2016.2573140
    [12] Li Y, Zhang B. An intrusion detection model based on multi-scale CNN. In: Proceedings of the 3rd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC). Chengdu, China: IEEE, 2019. 214−218
    [13] Lin W H, Lin H C, Wang P, Wu B H, Tsai J Y. Using convolutional neural networks to network intrusion detection for cyber threats. In: Proceedings of the IEEE International Conference on Applied System Invention (ICASI). Chiba, Japan: IEEE, 2018. 1107−1110
    [14] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. In: Proceedings of the 3rd International Conference on Learning Representations (ICLR). San Diego, USA: ICLR, 2015. 1−14
    [15] Szegedy C, Liu W, Jia Y Q, Sermanet P, Reed S, Anguelov D, et al. Going deeper with convolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston, USA: IEEE, 2015. 1−9
    [16] He K M, Zhang X Y, Ren S Q, Sun J. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, USA: IEEE, 2016. 770−778
    [17] Yang S. Research on network behavior anomaly analysis based on bidirectional LSTM. In: Proceedings of the 3rd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC). Chengdu, China: IEEE, 2019. 798−802
    [18] Hossain D, Ochiai H, Fall D, Kadobayashi Y. LSTM-based network attack detection: Performance comparison by hyper-parameter values tuning. In: Proceedings of the 7th IEEE International Conference on Cyber Security and Cloud Computing (CSCloud)/the 6th IEEE International Conference on Edge Computing and Scalable Cloud (EdgeCom). New York, USA: IEEE, 2020. 62−69
    [19] 陈红松, 陈京九. 基于循环神经网络的无线网络入侵检测分类模型构建与优化研究. 电子与信息学报, 2019, 41(6): 1427-1433

    Chen Hong-Song, Chen Jing-Jiu. Recurrent neural networks based wireless network intrusion detection and classification model construction and optimization. Journal of Electronics & Information Technology, 2019, 41(6): 1427-1433
    [20] Studer L, Wallau J, Ingold R, Fischer A. Effects of graph pooling layers on classification with graph neural networks. In: Proceedings of the 7th Swiss Conference on Data Science (SDS). Luzern, Switzerland: IEEE, 2020. 57−58
    [21] Hamilton W L, Ying Z, Leskovec J. Inductive representation learning on large graphs. In: Proceedings of the 31st Annual Conference on Neural Information Processing Systems. Long Beach, USA: Curran Associates Inc., 2017. 1025−1035
    [22] Chaudhary A, Mittal H, Arora A. Anomaly detection using graph neural networks. In: Proceedings of the International Conference on Machine Learning, Big Data, Cloud and Parallel Computing (COMITCon). Faridabad, India: IEEE, 2019. 346−350
    [23] 刘颖, 雷研博, 范九伦, 王富平, 公衍超, 田奇. 基于小样本学习的图像分类技术综述. 自动化学报, 2021, 47(2): 297-315

    Liu Ying, Lei Yan-Bo, Fan Jiu-Lun, Wang Fu-Ping, Gong Yan-Chao, Tian Qi. Survey on image classification technology based on small sample learning. Acta Automatica Sinica, 2021, 47(2): 297-315
    [24] Li Q Y, Shang Y L, Qiao X Q, Dai W. Heterogeneous dynamic graph attention network. In: Proceedings of the IEEE International Conference on Knowledge Graph (ICKG). Nanjing, China: IEEE, 2020. 404−411
    [25] Shanthamallu U S, Thiagarajan J J, Spanias A. A regularized attention mechanism for graph attention networks. In: Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Barcelona, Spain: IEEE, 2020. 3372−3376
    [26] Avelar P H C, Tavares A R, da Silveira T L T, Jung C R, Lamb L C. Superpixel image classification with graph attention networks. In: Proceedings of the 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI). Porto de Galinhas, Brazil: IEEE, 2020. 203−209
    [27] Vinayakumar R, Alazab M, Soman K P, Poornachandran P, Al-Nemrat A, Venkatraman S. Deep learning approach for intelligent intrusion detection system. IEEE Access, 2019, 7: 41525-41550 doi: 10.1109/ACCESS.2019.2895334
    [28] 杨印根, 王忠洋. 基于深度神经网络的入侵检测技术. 网络安全技术与应用, 2019(4): 37-41

    Yang Yin-Gen, Wang Zhong-Yang. Intrusion detection technology based on deep neural network. Network Security Technology & Application, 2019(4): 37-41
    [29] Shone N, Ngoc T N, Phai V D, Shi Q. A deep learning approach to network intrusion detection. IEEE Transactions on Emerging Topics in Computational Intelligence, 2018, 2(1): 41-50 doi: 10.1109/TETCI.2017.2772792
    [30] Liang W, Li K C, Long J, Kui X Y, Zomaya A Y. An industrial network intrusion detection algorithm based on multifeature data clustering optimization model. IEEE Transactions on Industrial Informatics, 2020, 16(3): 2063-2071 doi: 10.1109/TII.2019.2946791
    [31] Huang S K, Lei K. IGAN-IDS: An imbalanced generative adversarial network towards intrusion detection system in ad-hoc networks. Ad Hoc Networks, 2020, 105: 102177 doi: 10.1016/j.adhoc.2020.102177
    [32] Kozik R, Choraś M, Ficco M, Palmieri F. A scalable distributed machine learning approach for attack detection in edge computing environments. Journal of Parallel and Distributed Computing, 2018, 119: 18-26 doi: 10.1016/j.jpdc.2018.03.006
    [33] Prabavathy S, Sundarakantham K, Shalinie S M. Design of cognitive fog computing for intrusion detection in Internet of Things. Journal of Communications and Networks, 2018, 20(3): 291-298 doi: 10.1109/JCN.2018.000041
    [34] Fu Y S, Lou F, Meng F Z, Tian Z H, Zhang H, Jiang F. An intelligent network attack detection method based on RNN. In: Proceedings of the 3rd IEEE International Conference on Data Science in Cyberspace (DSC). Guangzhou, China: IEEE, 2018. 483−489
    [35] Khan F A, Gumaei A, Derhab A, Hussain A. A novel two-stage deep learning model for efficient network intrusion detection. IEEE Access, 2019, 7: 30373-30385 doi: 10.1109/ACCESS.2019.2899721
    [36] Elsayed M S, Le-Khac N A, Dev S, Jurcut A D. DDoSNet: A deep-learning model for detecting network attacks. In: Proceedings of the 21st International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM). Cork, Ireland: IEEE, 2020. 391−396
    [37] 陈晋音, 章燕, 王雪柯, 蔡鸿斌, 王珏, 纪守领. 深度强化学习的攻防与安全性分析综述. 自动化学报, 2022, 48(1): 21-39

    Chen Jin-Yin, Zhang Yan, Wang Xue-Ke, Cai Hong-Bin, Wang Jue, Ji Shou-Ling. A survey of attack, defense and related security analysis for deep reinforcement learning. Acta Automatica Sinica, 2022, 48(1): 21-39
    [38] Suwannalai E, Polprasert C. Network intrusion detection systems using adversarial reinforcement learning with deep Q-network. In: Proceedings of the 18th International Conference on ICT and Knowledge Engineering (ICT&KE). Bangkok, Thailand: IEEE, 2020. 1−7
  • 加载中
图(15) / 表(10)
计量
  • 文章访问数:  1479
  • HTML全文浏览量:  517
  • PDF下载量:  130
  • 被引次数: 0
出版历程
  • 收稿日期:  2020-10-01
  • 录用日期:  2021-01-19
  • 网络出版日期:  2021-03-27
  • 刊出日期:  2023-07-20

目录

    /

    返回文章
    返回