Metaheuristic-Driven Optimization for Efficient Resource Allocation in Cloud Environments

International Journal of Computational and Experimental Science and Engineering, Journal Year: 2025, Volume and Issue: 11(1)

Published: Jan. 7, 2025

Intrusion Detection Systems (IDS) play a pivotal role in safeguarding networks against evolving cyber threats. This research focuses on enhancing the performance of IDS using deep learning models, specifically XAI, LSTM, CNN, and GRU, evaluated NSL-KDD dataset. The dataset addresses limitations earlier benchmarks by eliminating redundancies balancing classes. A robust preprocessing pipeline, including normalization, one-hot encoding, feature selection, was employed to optimize model inputs. Performance metrics such as Precision, Recall, F1-Score, Accuracy were used evaluate models across five attack categories: DoS, Probe, R2L, U2R, Normal. Results indicate that XAI consistently outperformed other achieving highest accuracy (91.2%) Precision (91.5%) post-BAT optimization. Comparative analyses confusion matrices protocol distributions revealed dominance DoS attacks highlighted specific challenges with R2L U2R study demonstrates effectiveness optimized detecting complex attacks, paving way for adaptive solutions.

Language: Английский

---

Lightweight CNN-BiLSTM based Intrusion Detection Systems for Resource-Constrained IoT Devices

2022 International Wireless Communications and Mobile Computing (IWCMC), Journal Year: 2024, Volume and Issue: unknown, P. 1558 - 1563

Published: May 27, 2024

Language: Английский

---

Optimizing Smart Home Intrusion Detection With Harmony-Enhanced Extra Trees

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 117761 - 117786

Published: Jan. 1, 2024

In this study, we present an innovative network intrusion detection system (IDS) tailored for Internet of Things (IoT)-based smart home environments, offering a novel deployment scheme that addresses the full spectrum security challenges. Distinct from existing approaches, our comprehensive strategy not only proposes model but also incorporates IoT devices as potential vectors in cyber threat landscape, consideration often neglected previous research. Utilizing harmony search algorithm (HSA), refined extra trees classifier (ETC) by optimizing extensive array hyperparameters, achieving level sophistication and performance enhancement surpasses typical methodologies. Our was rigorously evaluated using robust real-time dataset, uniquely gathered 105 devices, reflecting more authentic complex scenario compared to simulated or limited datasets prevalent literature. commitment collaborative progress cybersecurity is demonstrated through public release source code. The underwent exhaustive testing 2-class, 8-class, 34-class configurations, showcasing superior accuracy (99.87%, 99.51%, 99.49%), precision (97.41%, 96.02%, 96.07%), recall (98.45%, 87.14%, 87.1%), f1-scores (97.92%, 90.65%, 90.61%) firmly establish its efficacy. Thiswork marks significant advancement security, providing scalable effective IDS solution adaptable intricate dynamics modern networks. findings pave way future endeavors realm defense, ensuring homes remain safe havens era digital vulnerability.

Language: Английский

---

Enhanced feature selection and ensemble learning for cardiovascular disease prediction: hybrid GOL2-2 T and adaptive boosted decision fusion with babysitting refinement

Frontiers in Medicine, Journal Year: 2024, Volume and Issue: 11

Published: July 5, 2024

Global Cardiovascular disease (CVD) is still one of the leading causes death and requires enhancement diagnostic methods for effective detection early signs prediction outcomes. The current tools are cumbersome imprecise especially with complex diseases, thus emphasizing incorporation new machine learning applications in differential diagnosis.

Language: Английский

---

A deep learning-based approach with two-step minority classes prediction for intrusion detection in Internet of Things networks

Knowledge-Based Systems, Journal Year: 2025, Volume and Issue: unknown, P. 113143 - 113143

Published: Feb. 1, 2025

Language: Английский

---

Generalizability Assessment of Learning‐Based Intrusion Detection Systems for IoT Security: Perspectives of Data Diversity

Security and Privacy, Journal Year: 2025, Volume and Issue: 8(2)

Published: March 1, 2025

ABSTRACT Machine learning (ML) and deep (DL) models have become vital tools in Intrusion Detection Systems (IDS), yet their effectiveness depends heavily on the quality distribution of training data. This study investigates impact dataset size balance performance ML DL using CIC‐IDS 2017 dataset. Five subsets (20%, 40%, 60%, 80%, 100% dataset) were created to assess across varying sizes. Four models, including Random Forest (RF), Artificial Neural Network, Convolutional Network (CNN), CNN+Long‐Term Short Memory (CNN+LSTM), trained evaluated these subsets, focusing precision, recall, F1‐score. To test model generalizability, a synthetic 20 million over‐sampled samples was generated Synthetic Minority Oversampling Technique, followed by manual under‐sampling create balanced 1.5 with approximately 100 000 per attack class. Upon generalizability assessment already synthetically datasets, CNN+LSTM consistently outperformed other but utilized more time for testing each case. The RF showed weakest performances fastest both scenarios. Moreover, evaluate importance general particular, we also considered NSL‐KDD all four multiple classifications binary classification. Our results highlight dataset, structure models.

Language: Английский

---

NSGTO‐LSTM: Niche‐strategy‐based gorilla troops optimization and long short‐term memory network intrusion detection model

ETRI Journal, Journal Year: 2025, Volume and Issue: unknown

Published: March 17, 2025

Abstract In recent decades, the rapid growth of Internet Things (IoT) has highlighted several network security problems. this study, an efficient intrusion detection (ID) system is implemented by using both machine learning and data mining concepts for detecting patterns. During initial phase, are collected from NSL‐KDD University New South Wales‐Network Based 15 (UNSW‐NB15) datasets. The then normalized/scaled employing a standard scaler technique. Next, informative feature values selected proposed optimization algorithm—that is, Niche‐Strategy‐based Gorilla Troops Optimization (NSGTO) algorithm. Finally, these transferred to Long Short‐Term Memory (LSTM) model classify types attacks on comparison existing ID systems, based NSGTO‐LSTM obtains classification accuracy 99.98% 99.90%

Language: Английский

---

Federated learning with LSTM for intrusion detection in IoT-based wireless sensor networks: a multi-dataset analysis

PeerJ Computer Science, Journal Year: 2025, Volume and Issue: 11, P. e2751 - e2751

Published: March 28, 2025

Intrusion detection in Internet of Things (IoT)-based wireless sensor networks (WSNs) is essential due to their widespread use and inherent vulnerability security breaches. Traditional centralized intrusion systems (IDS) face significant challenges data privacy, computational efficiency, scalability, particularly resource-constrained IoT environments. This study aims create assess a federated learning (FL) framework that integrates with long short-term memory (LSTM) for efficient IoT-based WSNs. We design the enhance accuracy, minimize false positive rates (FPR), ensure while maintaining system scalability. Using an FL approach, multiple nodes collaboratively train global LSTM model without exchanging raw data, thereby addressing privacy concerns improving capabilities. The proposed was tested on three widely used datasets: WSN-DS, CIC-IDS-2017, UNSW-NB15. evaluation metrics its performance included F1 score, FPR, root mean square error (RMSE). evaluated FL-based against traditional models, finding improvements detection. achieved higher accuracy lower FPR across all datasets than models. It effectively managed sequential WSNs, ensuring competitive performance, complex attack scenarios. work well together make strong way find intrusions which improves both underscores potential address key security, including making suitable real-world applications.

Language: Английский

---

Hybrid Compression: Integrating Pruning and Quantization for Optimized Neural Networks

Communications in computer and information science, Journal Year: 2025, Volume and Issue: unknown, P. 54 - 64

Published: Jan. 1, 2025

Language: Английский

---

Adaptive Intrusion Mitigation in Software-Defined Vehicles Using Deep Reinforcement Learning

Published: April 23, 2025

Language: Английский

---