Feature Selection in the Diabetes Dataset with the Marine Predator Algorithm and Classification using Machine Learning Methods DOI Creative Commons
F. J. Turk, Nuri Alper METİN, Zahide Nur YILMAZ

et al.

Gazi Üniversitesi Fen Bilimleri Dergisi Part C Tasarım ve Teknoloji, Journal Year: 2024, Volume and Issue: 12(3), P. 746 - 757

Published: Sept. 27, 2024

Diabetes, which is classified as one of the leading causes mortality, a chronic and intricate metabolic disorder defined by disruptions in metabolism carbohydrates, fats, proteins. Type 1 diabetes categorized alongside 2 diabetes, well other distinct kinds including gestational diabetes. Complications, both acute chronic, manifest individuals with due to diminished insulin secretion Following completion data preparation step, dataset that was collected from Kaggle then sent feature extraction module for analysis. After optimization process has been completed, selection block will determine characteristics stand out most. The selected traits discussed before are sorted into several categories using categorization module. findings compared those would have obtained if marine predator algorithm (MPOA) technique had not carried out, specifically regarding metrics like F1 score, Recall, Accuracy, Precision. indicate LR classification approach achieves an accuracy rate 77.63% without property selection. However, when MPOA, increases 79.39%.

Language: Английский

Analysis of Marine Predators Algorithm using BIAS toolbox and Generalized Signature Test DOI Creative Commons

Manish Kumar,

Kanchan Rajwar, Kusum Deep

et al.

Alexandria Engineering Journal, Journal Year: 2024, Volume and Issue: 95, P. 38 - 49

Published: March 29, 2024

The Marine Predators Algorithm (MPA) is a prominent Nature-Inspired Optimization (NIOA) that has garnered significant research interest due to its effectiveness. It draws inspiration from the foraging behaviors of marine predators, predominantly using Lévy or Brownian approach for strategy. Despite acclaim, structural bias within MPA not been thoroughly investigated, marking gap in current research. This absence targeted forms core rationale behind initiating this study. Structural recently identified NIOAs, causing population revisit specific regions search space without gaining new information. As result, it may lead increased computational costs and slow down rate convergence. Therefore, identifying essential better understand mechanism MPA. To ascertain presence any bias, two introduced models are employed: BIAS toolbox Generalized Signature Test. These examinations reveal notable MPA, towards center space. Also, possible future directions discussed. Our findings provide valuable insights into dynamics algorithm, fostering development new, unbiased, efficient algorithms.

Language: Английский

Citations

6
Improved marine predators algorithm for engineering design optimization problems DOI Creative Commons
Ye Chun,

Hua Xu,

Qi Chen

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: June 6, 2024

Abstract The Marine Predator Algorithm (MPA) has unique advantages as an important branch of population-based algorithms. However, it emerges more disadvantages gradually, such traps to local optima, insufficient diversity, and premature convergence, when dealing with complex problems in practical industrial engineering design applications. In response these limitations, this paper proposes a novel Improved (IMPA). By introducing adaptive weight adjustment strategy dynamic social learning mechanism, study significantly improves the encounter frequency efficiency between predators preys marine ecosystems. performance IMPA was evaluated through benchmark functions, CEC2021 suite problems, including welded beam design, tension/compression spring pressure vessel three-bar design. results indicate that achieved significant success optimization process over other methods, exhibiting excellent both solving optimal parameter solutions optimizing objective function values. performs well terms accuracy robustness, which also proves its successfully problems.

Language: Английский

Citations

4
Atom search optimization: a systematic review of current variants and applications DOI
Sylvère Mugemanyi, Zhaoyang Qu, François Xavier Rugema

et al.

Knowledge and Information Systems, Journal Year: 2025, Volume and Issue: unknown

Published: April 12, 2025

Language: Английский

Citations

0
Enhanced electricity price forecasting in smart grids using an optimized hybrid convolutional Multi-Layer Perceptron deep network with Marine Predators Algorithm for feature selection DOI
Bakir Abderrahim,

Abdelkader Rami

Energy Sources Part B Economics Planning and Policy, Journal Year: 2025, Volume and Issue: 20(1)

Published: Jan. 31, 2025

Language: Английский

Citations

0
Introduction to optimization techniques commonly used in materials science DOI
Sunil Kumar, Harbinder Singh, Simrandeep Singh

et al.

Elsevier eBooks, Journal Year: 2025, Volume and Issue: unknown, P. 131 - 168

Published: Jan. 1, 2025

Language: Английский

Citations

0
Fractional-order Izhikevich neuron Model: PI-rules numerical simulations and parameter identification DOI
Amr M. AbdelAty, Mohammed E. Fouda

Chaos Solitons & Fractals, Journal Year: 2025, Volume and Issue: 194, P. 116203 - 116203

Published: March 5, 2025

Language: Английский

Citations

0
NARX modeling and simulation of heave dynamics with application of robust control of an underactuated underwater vehicle DOI
S.M. Ahmad, Ahsan Tanveer

Ocean Engineering, Journal Year: 2025, Volume and Issue: 325, P. 120790 - 120790

Published: March 6, 2025

Language: Английский

Citations

0
Optimizing Intrusion Detection in Wireless Sensor Networks via the Improved Chameleon Swarm Algorithm for Feature Selection DOI Creative Commons
Laith Abualigah,

Mohammad H. Almomani,

Saleh Ali Alomari

et al.

IET Communications, Journal Year: 2025, Volume and Issue: 19(1)

Published: Jan. 1, 2025

ABSTRACT In this paper, the improved chameleon swarm algorithm (ICSA) enhances exploration–exploitation balance while optimizing feature subset selection. The integration of Lévy flight‐based exploration refines ICSA's search strategy, complemented by rotation‐type refinement and adaptive parameter‐setting mechanisms. These modifications ensure that aligns effectively with selection process, leading to a more efficient approach. To evaluate effectiveness, it is tested on NSL‐KDD benchmark, well‐established dataset in intrusion detection systems. Performance assessed based key metrics, including accuracy, rate, false alarm execution time, number selected features. Comparative analysis against six advanced classifiers demonstrates ICSA achieves superior results minimal computational overhead. attains highest accuracy (97.91%) rate (98.75%), fastest lowest (0.0021), eliminating need for excessive confirm modifying mechanisms within significantly efficiency performance, as validated through rigorous experimental testing at classifier level.

Language: Английский

Citations

0
Evolutionary Cost Analysis and Computational Intelligence for Energy Efficiency in Internet of Things-Enabled Smart Cities: Multi-Sensor Data Fusion and Resilience to Link and Device Failures DOI Creative Commons
Khalid A. Darabkh,

Muna Al-Akhras

Smart Cities, Journal Year: 2025, Volume and Issue: 8(2), P. 64 - 64

Published: April 9, 2025

This work presents an innovative, energy-efficient IoT routing protocol that combines advanced data fusion grouping and strategies to effectively tackle the challenges of management in smart cities. Our employs hierarchical Data Fusion Head (DFH), relay DFHs, marine predators algorithm, latter which is a reliable metaheuristic algorithm incorporates fitness function optimizes parameters such as how closely Sensor Nodes (SNs) group (DFG) are gathered together, distance sink node, proximity SNs within group, remaining energy (RE), Average Scale Building Occlusions (ASBO), Primary DFH (PDFH) rotation frequency. A key innovation our approach introduction techniques minimize redundant transmissions enhance quality DFG. By consolidating from multiple using algorithms, reduces volume transmitted information, leading significant savings. supports both direct routing, where fused flow straight multi-hop PDF chosen based on influential cost considers RE, ASBO. Given proposed efficient failure recovery strategies, redundancy management, techniques, it enhances overall system resilience, thereby ensuring high performance even unforeseen circumstances. Thorough simulations comparative analysis reveal protocol’s superior across metrics, namely, network lifespan, consumption, throughput, average delay. When compared most recent relevant protocols, including Particle Swarm Optimization-based clustering (PSO-EEC), linearly decreasing inertia weight PSO (LDIWPSO), Optimized Fuzzy Clustering Algorithm (OFCA), Novel PSO-based Protocol (NPSOP), achieves very promising results. Specifically, extends lifespan by 299% over PSO-EEC, 264% LDIWPSO, 306% OFCA, 249% NPSOP. It also consumption 254% relative 247% against 253% The throughput improvements reach 67% 59% 53% 50% fusing optimizing sets new benchmark for DFG, offering robust solution diverse deployments.

Language: Английский

Citations

0
Machine Learning Models for Predicting Compressive Strength of Eco-Friendly Concrete with Copper Slag Aggregates DOI
Yaser Moodi, Naser Safaeian Hamzehkolaei, Iman Afshoon

et al.

Materials Today Communications, Journal Year: 2025, Volume and Issue: unknown, P. 112572 - 112572

Published: April 1, 2025

Language: Английский

Citations

0