FUZ-SMO: A fuzzy slime mould optimizer for mitigating false alarm rates in the classification of underwater datasets using deep convolutional neural networks DOI Creative Commons
Dong Zhang, Zhiyong Jiang, F. Mohammadzadeh

et al.

Heliyon, Journal Year: 2024, Volume and Issue: 10(7), P. e28681 - e28681

Published: March 28, 2024

Sonar sound datasets are of significant importance in the domains underwater surveillance and marine research as they enable experts to discern intricate patterns within depths water. Nevertheless, task classifying sonar continues pose challenges. In this study, we present a novel approach aimed at enhancing precision efficacy dataset classification. The integration deep long-short-term memory (DLSTM) convolutional neural networks (CNNs) models is employed order capitalize on their respective advantages while also utilizing distinctive feature engineering techniques achieve most favorable outcomes. Although DLSTM have demonstrated effectiveness tasks involving sequence classification, attaining optimal performance necessitates careful adjustment hyperparameters. While traditional methods such grid random search effective, frequently encounter challenges related computational inefficiencies. This study aims investigate unexplored capabilities fuzzy slime mould optimizer (FUZ-SMO) context LSTM hyperparameter tuning, with objective addressing existing gap area. Drawing inspiration from adaptive behavior exhibited by moulds, FUZ-SMO proposes optimization. amalgamated model, which combines CNN, LSTM, fuzzy, SMO, exhibits notable improvement classification accuracy, outperforming conventional architectures margin 2.142%. model not only demonstrates accelerated convergence milestones but displays resilience against overfitting tendencies.

Language: Английский

Development and Evaluation of Transformer-Based Basic Fighter Maneuver Decision-Support Scheme for Piloting During Within-Visual-Range Air Combat DOI Creative Commons
Yiqun Dong, Shanshan He, Yunmei Zhao

et al.

Aerospace, Journal Year: 2025, Volume and Issue: 12(2), P. 73 - 73

Published: Jan. 21, 2025

In within-visual-range (WVR) air combat, basic fighter maneuvers (BFMs) are widely used. Air combat engagement database (ACED) is a dedicated for researching WVR combat. Utilizing the data in ACED, Transformer-based BFM decision support scheme developed to enhance pilot’s making The proposed model significantly outperforms baseline long short-term memory (LSTM)-based accuracy. To augment interpretability of this approach, Shapley Additive Explanation (SHAP) analysis employed, exhibiting rationality model’s decisions. Furthermore, study establishes comprehensive framework evaluating performance, validated through utilization from ACED. application experiments shows that increases winning rate 30% 70%, average percentage tactical advantage time 4.81% 14.73%, and situational share 17.83% 25.19%, which substantially improves thereby validating its effectiveness applicability scenarios.

Language: Английский

Citations

0
Threat Prediction Method for Large-Scale Beyond-Visual-Range Air Confrontation DOI

Li Leyan,

Lv Maolong,

Ao Wu

et al.

Lecture notes in electrical engineering, Journal Year: 2025, Volume and Issue: unknown, P. 34 - 42

Published: Jan. 1, 2025

Language: Английский

Citations

0
Improved Evidential Three-Way Decisions in Incomplete Multi-Scale Information Systems DOI
Rui Li, Chao Zhang, Deyu Li

et al.

International Journal of Approximate Reasoning, Journal Year: 2025, Volume and Issue: unknown, P. 109417 - 109417

Published: March 1, 2025

Language: Английский

Citations

0
A novel air combat target threat assessment method based on three-way decision and game theory under multi-criteria decision-making environment DOI
Qihong Chen, Qingsong Zhao, Zhigang Zou

et al.

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: 259, P. 125322 - 125322

Published: Sept. 6, 2024

Language: Английский

Citations

3
Exploring analytical solutions and modulation instability for the nonlinear fractional Gilson–Pickering equation DOI Creative Commons
Riaz Ur Rahman, Muhammad Bilal Riaz, Jan Martinovič

et al.

Results in Physics, Journal Year: 2024, Volume and Issue: 57, P. 107385 - 107385

Published: Feb. 1, 2024

The primary goal of this research is to explore the complex dynamics wave propagation as described by nonlinear fractional Gilson-Pickering equation (fGPE), a pivotal model in plasma physics and crystal lattice theory. Two alternative derivatives, termed β M-truncated, are employed analysis. new auxiliary method (NAEM) applied create diverse explicit solutions for surface waves given equation. This study includes comparative evaluation these using different types derivatives. derived fGPE, which include unique forms like dark, bright, periodic solitary waves, visually represented through 3D 2D graphs. These visualizations highlight shapes behaviors solutions, indicating significant implications industry innovation. proposed method's ability provide analytical demonstrates its effectiveness reliability analyzing models across various scientific technical domains. A comprehensive sensitivity analysis conducted on dynamical system fGPE. Additionally, modulation instability used assess model's stability, confirming robustness. verifies stability accuracy all solutions.

Language: Английский

Citations

2
A Novel Training Approach in Deep Spiking Neural Network Based on Fuzzy Weighting and Meta-heuristic Algorithm DOI Creative Commons
Melika Hamian, Karim Faez, Soheila Nazari

et al.

International Journal of Computational Intelligence Systems, Journal Year: 2024, Volume and Issue: 17(1)

Published: Feb. 19, 2024

Abstract The challenge of supervised learning in spiking neural networks (SNNs) for digit classification from speech signals is examined this study. Meta-heuristic algorithms and a fuzzy logic framework are used to train SNNs. Using gray wolf optimization (GWO), the features obtained audio reduced depending on dispersion each feature. Then, it combines weighting system (FWS) spike time-dependent flexibility (STDP) approach implement rule SNN. FWS produces uniformly distributed random weight STDP window, so that requires fewer training parameters. Finally, these neurons fed data estimate weights threshold values using wild horse algorithm (WHO). With parameters given, applied appropriately display class's share extracting relevant suggested network can classify into categories with 97.17% accuracy. dataset was operating at sparse biological rates below 600 Hz TIDIGITS test database. method has been evaluated IRIS Trip Data datasets, where results showed 98.93% 97.36% efficiency, respectively. Compared earlier efforts, study's demonstrate strategy both computationally simpler more accurate. accuracy digits, increased by 4.9, 3.46 1.24%, principal goal research improve SNN developing new high-precision method.

Language: Английский

Citations

2
A New Multi-Target Three-Way Threat Assessment Method with Heterogeneous Information and Attribute Relevance DOI Creative Commons
Yang Gao,

Na Lyu

Mathematics, Journal Year: 2024, Volume and Issue: 12(5), P. 691 - 691

Published: Feb. 27, 2024

Target threat assessment provides support for combat decision making. The multi-target method based on a three-way can obtain classification while receiving ranking, thus avoiding the limitation of traditional two-way decisions. However, heterogeneous situation information, attribute relevance, and adaptive information processing needs in complex battlefield environment bring challenges to existing methods. Therefore, this paper proposes new with relevance. Firstly, dynamic is represented by weights are calculated Criteria Importance Through Intercriteria Correlation (CRITIC). Then, conditional probability weighted Technique Order Preference Similarity Ideal Solution (TOPSIS), risk avoidance coefficients constructed calculating uncertainty value, then relative loss function matrices constructed. Finally, comprehensive obtained Heronian mean (HM) operator, thresholds rules. case study shows that compared methods, proposed effectively handle without presetting or field subjective settings, which more suitable mission environment.

Language: Английский

Citations

2
Novel reduction schemes for a dissipative dynamical system: A study on slow invariant manifolds in chemical kinetics DOI Creative Commons
Noureddine Elboughdiri, Faisal Sultan,

Muhammad Shoaib Ishaq

et al.

Ain Shams Engineering Journal, Journal Year: 2024, Volume and Issue: 15(6), P. 102751 - 102751

Published: March 13, 2024

This research explores the intricate concept of Slow Invariant Manifold (SIM) and its pivotal role in developing model reduction techniques (MRTs) for challenges within dissipative systems chemical kinetics, specifically mechanical engineering. Focusing on multi-step mechanism with two intermediates, primary approximations SIM are constructed compared using prominent MRTs: The Spectral Quasi Equilibrium (SQEM) Intrinsic Low Dimensional (ILDM). At given rate coefficient, a special computational experiment was performed which efficiency species has been compared. Noteworthy innovation involves evaluating separately reduced species, departing from conventional approach considering every mechanism. study employs local sensitivity analysis MATLAB's Sim-Biology toolbox, presenting quantitative findings tabular format comprehensive MRT comparison. Beyond contributing to deeper understanding complex this marks first exploration systems. novel perspective offers nuanced insights, emphasizing critical effectively addressing engineering applications. In summary, introduces advancements approaches, advancing highlighting significance contexts.

Language: Английский

Citations

2
Towards improving community detection in complex networks using influential nodes DOI

Henghe Zheng,

Haitao Zhao,

Gholamreza Ahmadi

et al.

Journal of Complex Networks, Journal Year: 2023, Volume and Issue: 12(1)

Published: Dec. 22, 2023

Abstract Complex network analysis is inspired by empirical studies of real-world networks such as computer networks, technology and social networks. The community structure in complex understood an important issue the research society. A a set nodes where density connections high. insight literature shows many approaches to identify influential nodes, but these only lead finding centres. Meanwhile, clustering techniques are effectively used for detection, they can reveal group hidden considering topological demographic information. This article presents ensemble algorithm based on improve detection Considering different characteristics network, proposed method seeks discover common interests between users their behaviours most suitable communities. First, identified Then, centres considered cluster After that, primary clusters created determined Finally, reclustered form final clusters. Here, communities network. simulation has been performed results confirm effectiveness method. Specifically, 2.1% better than best existing state-of-the-art terms modularity. Keywords: network; detection; nodes; clustering.

Language: Английский

Citations

6
Homogeneous–heterogeneous reactions in the colloidal investigation of Casson fluid DOI Creative Commons
Saima Noor

Open Physics, Journal Year: 2024, Volume and Issue: 22(1)

Published: Jan. 1, 2024

Abstract With particular attention to the effects of an electromagnetically induced resistive force on homogeneous–heterogeneous processes and related homogeneous heat effects, Casson fluid flow towards a stretching sheet at magnetohydrodynamic stagnation point is investigated in detail. In this situation, Laplace approach helps decipher subtleties first-order kinetics governing fluid’s motion. Notably, dynamics are largely determined by behaviour expected surrounding environment, forming strong correlation between catalyst temperature wall surface activity. Using conventional differential systems, our analysis gains great deal from modified decomposition method, which allows non-linear systems be computed examined. order improve understanding, numerical findings included, graphs skillfully used examine different factors. The in-depth examination also includes complicated patterns concentration temperature, providing insightful information intricate interactions forces dynamic system.

Language: Английский

Citations

1