Electrocardiogram (ECG) classification is a critical task for early detection of cardiac abnormalities.However, standard machine learning classifiers such as Logistic Regression (LR), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Random Forest (RF) face challenges when handling high-dimensional data.These classifiers often struggle to model complex relationships among features, leading to reduced classification performance.Existing metaheuristic-based feature selection methods typically search across the entire feature space, which increases computational complexity and often fails to isolate only the most informative features.This paper introduces a novel hybrid feature selection approach based on a two-phase framework combining Topsoe Distance (TD) and Particle Swarm Optimization (PSO).In the first phase, TD is used to eliminate irrelevant and redundant features, thereby reducing the feature space from 4000 to 1600 and preserving only the most relevant and harmonized features.In the second phase, PSO is applied to the reduced space, enabling efficient optimization and feature subset selection.Experimental results using the MIT-BIH Arrhythmia dataset demonstrate that the proposed TD-PSO method significantly improves classification accuracy compared to using PSO alone.Specifically, LR accuracy improved from 87% to 95%, KNN from 92% to 97%, SVM from 92% to 96%, and RF from 93% to 99%.These results confirm that narrowing the search space before optimization allows classifiers to perform more effectively while reducing computational cost.The proposed TD-PSO framework provides a robust and scalable solution for high-dimensional ECG classification and holds strong potential for broader applications in biomedical data analysis.

Coronary artery disease (CAD) is a major cause of death in the world, and requires accurate and rapid diagnosis.This study presents a new hybrid framework combining feature selection, hybrid classification, and Bayesian optimization for the improvement of CAD detection.The first approach screens the most relevant features using an Information Gain (IG) as filter feature selection method, and then those selected features are optimized by Grey Wolf Optimizer (GWO) algorithm as a wrapper feature selection method.For the classification stage, we utilized a hybrid ensemble of AdaBoost and XGBoost to exploit their complementary strengths and increase diagnostic performance.In addition, Bayesian Optimization is implemented to perform hyperparameters tuning for AdaBoost to produce the best model.Experimental results show that the proposed method achieves an average recognition accuracy of 98.72% on the Z-Alizadeh Sani dataset with the K-fold cross-validation technique, which performs much better than the conventional methods that have used the Z-Alizadeh Sani dataset.This hybrid approach has a potential of classification flexibility as a result of its feasibility and a minimum overfitting error, and an efficient selection of features, making it a reliable solution for CAD diagnosis.

Cloud datacenters form the underlying infrastructure of cloud computing, containing a large number of physical machines (servers).The servers are virtualized to host many virtual machines (VMs) that support a wide range of services for cloud users.Efficient management of VMs is crucial to achieve optimum resource utilization, better energy efficiency, and a high degree of quality of service (QoS).Virtual machines consolidation is one of the main mechanisms to achieve such objectives.Consolidation techniques involve dynamically re-allocation of VMs across physical machines to balance workloads, reduce energy consumption, and maintain system reliability.However, VM consolidation is a complex process that can introduce significant overhead in terms of power consumption, network bandwidth, and service disruption.This makes its optimization a critical challenge for cloud service providers.This paper proposes DHLT-ACO (Dynamic Host Load Thresholding with Ant Colony Optimization) to enhance VM consolidation through two key mechanisms.At first, an adaptive host load detection method based on exponentially weighted moving average (EWMA) and a confidence interval-based dynamic thresholds to enhance the accuracy of detecting underloaded and overloaded hosts.Second, an improved Ant Colony Optimization (ACO) algorithm to optimize VMs placement by dynamically adjusting exploration and exploitation parameters, ensuring faster convergence and better resource allocation.Experimental results demonstrate that DHLT-ACO outperforms existing methods in terms of power consumption, number of VMs migration, and minimizing service level agreement (SLA) violations.DHLT-ACO achieved energy savings of 21.3%, reduced the number of migrations by 47%, and minimized SLA violations by 45.1% over the utilization-aware VM placement (UAVMP), hybrid VM allocation and placement (HVMAP), normalization-based VM consolidation (NVMC), and modified genetic-based VM consolidation (MGVMC) on average.These improvements verify the efficiency of DHLT-ACO in achieving energy efficiency, reducing the number of migrations and ensuring compliance with SLA in cloud computing environments.