Horse Herd optimization with deep learning based intrusion detection in cloud computing environment

• Published in: International journal of information technology (Singapore. Online) Vol. 17; no. 1; pp. 387 - 393
• Main Authors: Nagamani, Samineni, Arivalagan, S., Senthil, M., Sudhakar, P.
• Format: Journal Article
• Language: English
• Published: Singapore Springer Nature Singapore 01.01.2025; Springer Nature B.V
• Subjects: Accuracy; Artificial Intelligence; Classification; Cloud computing; Communications traffic; Computer Imaging; Computer Science; Cybersecurity; Datasets; Deep learning; Deer; Feature selection; Image Processing and Computer Vision; Intrusion detection systems; Machine Learning; Neural networks; Optimization; Optimization algorithms; Original Research; Pattern Recognition and Graphics; Recurrent neural networks; Software; Software Engineering; Support vector machines; Vision; Deep learning; Cloud computing; Intrusion detection system; Data normalization; Horse herd optimization
• ISSN: 2511-2104; 2511-2112
• DOI: 10.1007/s41870-024-02199-w

The cloud offers applications, infrastructure, and storage services to consumers that must be secure by some strategies. Hence, security in the cloud is to protect consumer data and structure from malicious users by delivering integrity, availability, confidentiality, and in-time intrusion recognition. Utilizing deep learning (DL), intrusion detection systems (IDS) employ advanced neural networks to automatically recognize and respond to fraudulent activities,. By analyzing large-scale datasets of network traffic, DL techniques like Recurrent Neural Network (RNN) and Long Short-Term Memory network (LSTM), can distinguish patterns linked with several cyber-attacks. This study presents a novel Horse Herd Optimization with a Deep Learning based Intrusion Detection Approach (HHODL-IDA) methodology in Cloud Computing. The goal of the HHODL-IDA methodology is to achieve security in the cloud platform by employing intrusion detection. In the HHODL-IDA technique, min-max scalar is primarily utilized to scale the input data. To select the features, the HHODL-IDA technique involves the invasive weed optimization (IWO) technique. Next, the detection of intrusions takes place using attention-based bidirectional LSTM (A-BiLSTM) technique. Eventually, the HHO approach has been executed for the enhanced hyperparameter selection of the A-BiLSTM approach. The experimental value of the HHODL-IDA approach has been executed using a benchmark IDS database. The extensive comparison study stated that the HHODL-IDA approach outcomes in greater detection results in the CC platform.