Meta-Learner-Based Approach for Detecting Attacks on Internet of Things Networks

Sensors, Journal Year: 2023, Volume and Issue: 23(19), P. 8191 - 8191

Published: Sept. 30, 2023

The significant surge in Internet of Things (IoT) devices presents substantial challenges to network security. Hackers are afforded a larger attack surface exploit as more become interconnected. Furthermore, the sheer volume data these generate can overwhelm conventional security systems, compromising their detection capabilities. To address posed by increasing number interconnected IoT and overload they generate, this paper an approach based on meta-learning principles identify attacks within networks. proposed constructs meta-learner model stacking predictions three Deep-Learning (DL) models: RNN, LSTM, CNN. Subsequently, identification relies various methods, namely Logistic Regression (LR), Multilayer Perceptron (MLP), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost). assess effectiveness approach, extensive evaluations conducted using dataset from 2020. XGBoost showcased outstanding performance, achieving highest accuracy (98.75%), precision (98.30%), F1-measure (98.53%), AUC-ROC (98.75%). On other hand, SVM exhibited recall (98.90%), representing slight improvement 0.14% over performance achieved XGBoost.

Language: Английский

---

Majority Voting Ensemble Classifier for Detecting Keylogging Attack on Internet of Things

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 19860 - 19871

Published: Jan. 1, 2024

An intrusion attack on the Internet of Things (IoT) is any malicious activity or unauthorized access that jeopardizes integrity and security IoT systems, networks, devices. Regarding IoT, intrusions can result in severe problems, including service disruption, data theft, privacy violations, even bodily injury. One attacks a keylogging attack, sometimes referred to as keystroke logging keyboard capture, which type cyberattack attacker secretly observes records keystrokes made device's keyboard. In context where connected objects communicate exchange data, this assault may be especially concerning. Keylogging have repercussions ecosystem since they compromise sensitive information, login passwords, personal financial confidential communications. This paper explored possibility using an ensemble classifier detect networks. We built consisting three classifiers: convolutional neural network (CNN), recurrent (RNN), long-short memory (LSTM). A proposed model uses BoT-IoT dataset attack. Results show improve model's performance. The had excellent accuracy low false positive rate. It also significantly improved detection rates for than other classifiers.

Language: Английский

---

Novel machine learning investigation for Buongiorno fluidic model with Sutterby nanomaterial

Tribology International, Journal Year: 2024, Volume and Issue: 199, P. 110009 - 110009

Published: July 19, 2024

Language: Английский

---

Towards Efficient Information Retrieval in Internet of Things Environments Via Machine Learning Approaches

Journal of The Institution of Engineers (India) Series B, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 17, 2024

Language: Английский

---

An effective deep learning model for spatial-temporal significant wave height prediction in the Atlantic hurricane area

Ocean Engineering, Journal Year: 2024, Volume and Issue: 317, P. 120083 - 120083

Published: Dec. 18, 2024

Language: Английский

---