Wireless Sensor Networks (WSNs) are increasingly vulnerable to sophisticated cyber threats, necessitating advanced intrusion detection systems (IDS) that balance high accuracy with interpretability. This paper presents a Knowledge-Improved Hybrid Deep Neural Network-Kolmogorov Arnold Network (DNN-KAN) Framework for intrusion detection in WSNs, integrating data-driven learning with domain-specific knowledge to enhance detection performance. The proposed framework preprocesses and merges multiple datasets (WSN, NSL-KDD, and CICIDS2017), extracts features using Principal Component Analysis (PCA), and constructs a knowledge graph to embed expert-defined rules via Graph Convolutional Networks (GCNs). The hybrid architecture employs a DNN for hierarchical feature extraction and replaces the fully connected layer with a KAN layer, leveraging the Kolmogorov-Arnold Theorem to decompose complex relationships into interpretable univariate functions. Evaluated on key metrics, the model achieves 99.87% accuracy, 99.85% precision, and a 0.9985 ROC-AUC score, significantly outperforming traditional DNNs, standalone KANs, and state-of-the-art models like CNN-LSTM and Transformer-based IDS. By incorporating domain knowledge into the loss function, the framework reduces false positives and improves generalization, even with limited training data. This work demonstrates that combining knowledge-driven constraints with hybrid neural architectures can yield highly accurate, interpretable, and robust intrusion detection for WSN security.
Data mining (DM) involves the process of identifying patterns, correlation, and anomalies existing in massive datasets. The applicability of DM includes several areas such as education, healthcare, business, and finance. Educational Data Mining (EDM) is an interdisciplinary domain which focuses on the applicability of DM, machine learning (ML), and statistical approaches for pattern recognition in massive quantities of educational data. This type of data suffers from the curse of dimensionality problems. Thus, feature selection (FS) approaches become essential. This study designs a Feature Subset Selection with an optimal machine learning model for Educational Data Mining (FSSML-EDM). The proposed method involves three major processes. At the initial stage, the presented FSSML-EDM model uses the Chicken Swarm Optimization-based Feature Selection (CSO-FS) technique for electing feature subsets. Next, an extreme learning machine (ELM) classifier is employed for the classification of educational data. Finally, the Artificial Hummingbird (AHB) algorithm is utilized for adjusting the parameters involved in the ELM model. The performance study revealed that FSSML-EDM model achieves better results compared with other models under several dimensions.
This current work provides a comprehensive Computational Fluid Dynamics (CFD) investigation of three-dimensional magnetohydrodynamic (MHD) mixed convection of carbon nanotube (CNT)-water nanofluid within a vented rectangular cavity featuring an internal vertical rod bundle with circular, square, and triangular cross-sections. The finite element method (FEM) was used to investigate the effects of key parameters, including the Richardson number (0.01 ≤ Ri ≤ 10), Hartmann number (0 ≤ Ha ≤ 100), and CNT nanoparticle concentration (0 ≤ ϕ ≤ 0.045), in relation to fluid flow and heat transfer performance. The CNT nanoparticle incorporation increases the nanofluid’s heat transfer capacity by up to 22%, with the highest average Nusselt number (Nuav) achieved with circular rods at ϕ = 0.045, which corresponds to the higher convective heat transfer efficiency. The magnetic field further stabilizes the flow by reducing thermal convection irregularities, with a 15% improvement in temperature distribution uniformity when Ha = 100. The investigation’s outcomes reveal that due to their smoother geometries, the circular rods exhibit better thermal exchange rates compared to square and triangular rods. Moreover, a polynomial regression model is used to correlate the governing parameters and heat transfer rates, and it achieves a high R2 of 0.964. These findings highlight the potential of CNT-water nanofluid and magnetic field applications for thermal management optimization in various engineering systems.
This numerical investigation explores the enhanced control of the 3D natural convection (NC) within a cubic cavity filled with graphene–water nanofluids, utilizing a bottom-center-located tree-shaped obstacle and a horizontal magnetic field (MF). The analysis includes the effects of the Rayleigh number (Ra), the solid volume fraction of graphene (), the Hartmann number (Ha), and the fins’ length (W). The results show complex flow patterns and thermal behavior within the cavity, indicating the interactive effects of nanofluid properties, the tree-shaped obstacle, and magnetic field effects. The MHD effects reduce the convection, while the addition of graphene improves the thermal conductivity of the fluid, which enhances the heat transfer observed with increasing Rayleigh numbers. The increase in the fins’ length on the heat transfer efficiency is found to be slightly negative, which is attributed to the complex interplay between the enhanced heat transfer surface area and fluid flow disruption. This study presents an original combination of non-destructive methods (magnetic field) and a destructive method (tree-shaped obstacle) for the control of the fluid flow and heat transfer characteristics in a 3D cavity filled with graphene–water nanofluids. In addition, it provides valuable information for optimizing heat transfer control strategies, with applications in electronic cooling, renewable energy systems, and advanced thermal management solutions. The application of a magnetic field was found to reduce the maximum velocity and total entropy generation by about 82% and 76%, respectively. The addition of graphene nanoparticles was found to reduce the maximum velocity by about 5.5% without the magnetic field and to increase it by 1.12% for Ha = 100. Varying the obstacles’ length from W = 0.2 to W = 0.8 led to a reduction in velocity by about 23.6%.