Performance Analysis of Different Activation and Loss Functions of Stacked Autoencoder for Dimension Reduction for NIDS on Cloud Environment

  IJETT-book-cover  International Journal of Engineering Trends and Technology (IJETT)          
  
© 2021 by IJETT Journal
Volume-69 Issue-4
Year of Publication : 2021
Authors : Nirmalajyothi Narisetty, Gangadhara Rao Kancherla, Basaveswararao Bobba, K. Swathi
DOI :  10.14445/22315381/IJETT-V69I4P224

Citation 

MLA Style: Nirmalajyothi Narisetty, Gangadhara Rao Kancherla, Basaveswararao Bobba, K. Swathi  "Performance Analysis of Different Activation and Loss Functions of Stacked Autoencoder for Dimension Reduction for NIDS on Cloud Environment" International Journal of Engineering Trends and Technology 69.4(2021):169-176. 

APA Style:Nirmalajyothi Narisetty, Gangadhara Rao Kancherla, Basaveswararao Bobba, K. Swathi. Performance Analysis of Different Activation and Loss Functions of Stacked Autoencoder for Dimension Reduction for NIDS on Cloud Environment  International Journal of Engineering Trends and Technology, 69(4),169-176.

Abstract
This paper`s objective is twofold, i) to provide a framework for a better intrusion detection system with an SVM classifier to detect new types of attacks in a cloud environment. ii) Performance comparative study is carried out to identify the best combination of Stacked Autoencoder (SAE) activation and loss functions for dimension reduction. To achieve the first objective, the CICIDS2017 is considered because it consists of modern attacks on Cloud environment-related. The Stacked Autoencoder with backpropagation and Adam Optimizer algorithms meet the second objective of this paper. For this purpose, to conducting experiments, three activation functions and two-loss functions are considered. The Activation functions Rectified Linear Unit (ReLU), SoftMax, and Scaled Exponential Linear Unit (SeLU) are being used as input/hidden and output layers. For loss functions, Mean Squared Error (MSE) and Cross-Entropy (CE) are chosen. To find the effect of these functions` performance metrics, accuracy, precision, recall, f-measure, and computational time are evaluated with SVM classifier using CICIDS 2017 benchmark dataset. The experimental results show that the ReLU-ReLU-CE yields better accuracy, and the SeLU-SeLU-MSE executes with less computational time.

Reference
[1] Deshmukh, R. V., & Devadkar, K. K., Understanding DDoS attack & its effect. In cloud environment., Procedia Computer Science, 49 202-210.
[2] Zekri, M., El Kafhali, S., Aboutabit, N., & Saadi, Y., DDoS attack detection using machine learning techniques in cloud computing environments, (2017).
[3] In 3rd International Conference of Cloud Computing Technologies and Applications (CloudTech) IEEE 1-7, (2017).
[4] Kumar, R., Lal, S. P., & Sharma, A., Detecting denial of service attacks in the cloud. In IEEE 14th Intl Conf on Dependable, Autonomic and Secure Computing, 14th Intl Conf on Pervasive Intelligence and Computing, 2nd Intl Conf on Big Data Intelligence and Computing and Cyber Science and Technology Congress (DASC/PiCom/DataCom/CyberSciTech) (2016) 309-316. IEEE.
[5] Abbasi, H., Ezzati-Jivan, N., Bellaiche, M., Talhi, C., & Dagenais, M. R., Machine learning-based EDoS attack detection technique using execution trace analysis, Journal of Hardware and Systems Security, 3(2)(2019) 164-176.
[6] Al-Qatf, M., Lasheng, Y., Al-Habib, M., & Al-Sabahi, K., Deep learning approach combining sparse autoencoder with SVM for network intrusion detection, IEEE Access, 6 (2019) 52843-52856.
[7] Kunang, Y. N., Nurmaini, S., Stiawan, D., Zarkasi, A., & Jasmir, F., Automatic Features Extraction Using Autoencoder in Intrusion Detection System. In International Conference on Electrical Engineering and Computer Science (ICECOS) (2018) 219-224. IEEE.
[8] Wang, W., Du, X., Shan, D., & Wang, N., A Hybrid Cloud Intrusion Detection Method Based on SDAE and SVM, In12th International Conference on Intelligent Computation Technology and Automation (ICICTA) (2019) 271-274. IEEE. DOI 10.1109/ICICTA49267.2019.00064
[9] Wang, Y., Li, Y., Song, Y., & Rong, X., The Influence of the Activation Function in a Convolution Neural Network Model of Facial Expression Recognition, Applied Sciences, 10(5)(2020) 1897.
[10] Isa, F. M., Buja, A. G., Darus, M. Y., & Saad, S., Optimizing the Effectiveness of Intrusion Detection System by using Pearson Correlation and Tune Model Hyper Parameter on Microsoft Azure Platform. International Journal, 9(1.3)(2020).
[11] Abdulrahman, A. A., & Ibrahem, M. K., Evaluation of DDoS attacks Detection in a New Intrusion Dataset Based on Classification Algorithms., Iraqi Journal of Information & Communications Technology, 1(3)(2018) 49-55.
[12] Abdulhammed, R., Musafer, H., Alessa, A., Faezipour, M., & Abuzneid, A., Features dimensionality reduction approaches for machine learning-based network intrusion detection., Electronics, 8(3) (2019) 322.
[13] Gupta, N., Bedi, P., & Jindal, V., Effect of Activation Functions on the Performance of Deep Learning Algorithms for Network Intrusion Detection Systems, In Proceedings of ICETIT (2019) 949-960. Springer, Cham.
[14] Panigrahi, R., & Borah, S.., A detailed analysis of CICIDS2017 dataset for designing Intrusion Detection Systems., International Journal of Engineering & Technology, 7(3.24)(2018) 479-482.
[15] Yousefi-Azar, M., Varadharajan, V., Hamey, L., & Tupakula, U. (2017)., Autoencoder-based feature learning for cybersecurity applications., In 2017 International joint conference on neural networks (IJCNN) 3854-3861. IEEE.
[16] Meng, Q., Catchpoole, D., Skillicom, D., & Kennedy, P. J. 2017, May)., Relational autoencoder for feature extraction., In 2017 International Joint Conference on Neural Networks (IJCNN) 364-371. IEEE.
[17] Chen, R. C., Cheng, K. F., Chen, Y. H., & Hsieh, C. F., Using rough set and support vector machine for network intrusion detection system. In 2009 First Asian Conference on Intelligent Information and Database Systems (465-470) IEEE.
[18] Wang, Y., Cai, W. D., & Wei, P. C., A deep learning approach for detecting malicious JavaScript code. Security and Communication Networks, 9(11)(2016) 1520-1534.
[19] POONGOTHAI, T., & JAYARAJAN, K., An Effective and Intelligent Intrusion Detection System using Deep Auto-Encoders.
[20] Sharafaldin, I., Lashkari, A. H., & Ghorbani, A. A., Toward generating a new intrusion detection dataset and intrusion traffic characterization., In ICISSP (2018) 108-116.
[21] ABDULRAHEEM, M. H., & IBRAHEEM, N. B., A DETAILED ANALYSIS OF NEW INTRUSION DETECTION DATASET., Journal of Theoretical and Applied Information Technology, 97(17)(2019).
[22] Krishna, K. V., Swathi, K., & Rao, B. B., A Novel Framework for NIDS through Fast kNN Classifier on CICIDS2017 Dataset.
[23] Banerjee, M., Mitra, S., & Anand, A., Feature selection using rough sets., In Multi-Objective Machine Learning (2006) 3-20. Springer, Berlin, Heidelberg.
[24] Caballero, Y., Alvarez, D., Bello, R., & Garcia, M. M., Feature selection algorithms using rough set theory., In Seventh International Conference on Intelligent Systems Design and Applications (ISDA 2007) 407-411. IEEE.
[25] Shen, Z., Chen, X., & Garibaldi, J., Performance optimization of a fuzzy entropy-based feature selection and classification framework., In 2018 IEEE International Conference on Systems, Man, and Cybernetics (SMC)(2018) 1361-1367. IEEE.
[26] Kingma, D. P., & Ba, J., Adam: A method for stochastic optimization., arXiv preprint arXiv:1412.6980., (2014).
[27] Ahmad Akl, Ahmed Moustafa, Ibrahim El-Henawy, Deep Learning: Approaches and Challenges, International Journal of Engineering Trends and Technology 65(1) (2018) 9-16.

Keywords
Auto-encoder, cloud computing, dimensionality reduction, intrusion detection system, machine learning