Article
Computer Science, Interdisciplinary Applications
Hang Su, Dong Zhao, Fanhua Yu, Ali Asghar Heidari, Zhangze Xu, Fahd S. Alotaibi, Majdi Mafarja, Huiling Chen
Summary: As science and technology advance, more complex engineering problems emerge, increasing the need for new optimization techniques. The cuckoo search algorithm has been widely used but can no longer meet current optimization requirements. This paper proposes an improved cuckoo search algorithm called CCFCS, which incorporates a crossover optimizer and a decentralized foraging strategy to enhance its search ability and ability to escape local optima. The algorithm's performance is verified through various tests and it shows faster convergence and higher solution quality compared to other algorithms.
JOURNAL OF COMPUTATIONAL DESIGN AND ENGINEERING
(2023)
Article
Automation & Control Systems
Seydali Ferahtia, Hegazy Rezk, Ali Djerioui, Azeddine Houari, Saad Motahhir, Samir Zeghlache
Summary: The Bald Eagle Search algorithm (BES) is a recent metaheuristic algorithm inspired by bald eagle hunting behavior. Adaptive parameters are introduced into the original BES to overcome its limitations and enhance exploration and exploitation. The modified BES is tested and compared with other algorithms, and analysis of variance and Tuckey tests are conducted to validate the results' significance.
Article
Computer Science, Artificial Intelligence
Yunhui Zhang, Yongquan Zhou, Guo Zhou, Qifang Luo
Summary: In this paper, a multi-objective bald eagle search algorithm (MOBES) is proposed, which introduces an archive mechanism and elite selection strategy to enhance efficiency. The MOBES outperforms its competitors in terms of convergence, diversity, and distribution of solutions on CEC 2020 benchmark functions. It is also proven to be more competitive in handling challenging multi-objective optimization problems in real-world engineering design.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Jiatang Cheng, Yan Xiong
Summary: MACS is an improved CS algorithm that enhances its versatility and robustness in solving complex optimization problems by employing parameter control strategy and integrating multiple search strategies.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Computer Science, Information Systems
Omar Rodriguez-Abreo, Jose Miguel Hernandez-Paredes, Alejandro Flores Rangel, Carlos Fuentes-Silva, Francisco Antonio Castillo Velasquez
Summary: The paper introduces a modified metaheuristic cuckoo search algorithm for parametric estimation of motors, using steady-state equations to determine parameters and comparing its performance with other algorithms, showing that it can calculate parameters more accurately.
Article
Computer Science, Artificial Intelligence
Hu Peng, Zhaogan Zeng, Changshou Deng, Zhijian Wu
Summary: Cuckoo search algorithm is effective but can get trapped in local optimum due to unitary search strategy. To overcome this, a multi-strategy serial CS algorithm (MSSCS) is proposed with new learning strategies based on cuckoo's behavior, aiming to enhance performance.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Engineering, Marine
Sheraz Aslam, Michalis P. Michaelides, Herodotos Herodotou
Summary: Maritime container terminals (MCTs) play a crucial role in international trade but face challenges such as congestion and high service costs. This study extends the berth allocation problem (BAP) to multiple quays and proposes the use of the cuckoo search algorithm (CSA) to solve the multi-quay BAP (MQ-BAP) by minimizing the total service cost. Experimental results show that the CSA-based method outperforms other metaheuristic approaches in terms of overall performance.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2023)
Article
Automation & Control Systems
Mohamed Siddiq Zatout, Amar Rezoug, Abdellah Rezoug, Khalifa Baizid, Jamshed Iqbal
Summary: This study applied three metaheuristic methods to optimize fuzzy logic controllers for quadrotor attitude stabilisation, and found that BAT algorithm outperformed PSO and CS in terms of performance, computation time, and fitness. The BAT-based fuzzy controller exhibited superior performance compared with other algorithms in stabilising the quadrotor.
INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE
(2022)
Article
Computer Science, Artificial Intelligence
Pedda Nagyalla Maddaiah, Pournami Pulinthanathu Narayanan
Summary: Artificial neural networks are widely used for solving engineering design problems due to their simplicity, efficiency, and adaptability. However, stochastic gradient descent methods suffer from issues such as vanishing gradient and sensitivity to initial weights and biases. The cuckoo search algorithm, a simple and efficient metaheuristic algorithm, can be used to overcome these problems.
NEURAL PROCESSING LETTERS
(2023)
Article
Computer Science, Artificial Intelligence
Rui Wu, Haisong Huang, Jianan Wei, Chi Ma, Yunwei Zhu, Yilin Chen, Qingsong Fan
Summary: This paper proposes the sparrow search algorithm based on quantum computations and multi-strategy enhancement (QMESSA) to address the issues of deficient optimization accuracy and low search efficiency in the traditional SSA. QMESSA utilizes a diversified initial population strategy combined with quantum computations and a quantum gate mutation mechanism to obtain a more random and diverse initial population. It also introduces an enhanced search strategy with an adaptive T-distribution and a new position update formula to accelerate convergence and enhance variability. Experimental results and statistical analysis demonstrate the superior performance of QMESSA over SSA and other advanced optimization algorithms. Furthermore, QMESSA exhibits superiority in solving classical practical application problems. The source code of QMESSA is available at https://ww2.mathworks.cn/matlabcentral/fileexchange/120013-project1-0.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Energy & Fuels
Bo Yang, Danyang Li, Chunyuan Zeng, Yiming Han, Junhui Li
Summary: Accurate and reliable parameter identification for proton exchange membrane fuel cells (PEMFC) is crucial for simulation analysis, optimal control, and performance research. Traditional optimization methods face difficulties in accurately and efficiently identifying parameters due to the strong coupling, inherent nonlinear, and multi-variable characteristics. In this study, an advanced bald eagle search (BES) algorithm is proposed to reliably identify the unknown parameters of the electrochemical PEMFC model. Results show that BES outperforms the genetic algorithm (GA) in parameter identification, achieving a 96.27% reduction in root mean square error (RMSE).
FRONTIERS IN ENERGY RESEARCH
(2022)
Article
Computer Science, Information Systems
Jiaxiang Zhang, Linwei Li, Huanlong Zhang, Fengxian Wang, Yangyang Tian
Summary: A novel sparrow search algorithm (NSSA) combined with spawning technology is proposed in this paper to effectively improve the problem of the algorithm falling into local optimization. The NSSA replaces the traditional stochastic method with the good point set theory to find the initial individual, integrates the spawning strategy of the cuckoo algorithm into the discoverer stage, and uses Levy flight and Brownian motion to disturb the position of the sparrow dimension by dimension. Simulation results show the effectiveness and superiority of the proposed scheme.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Xiaobing Yu, Wenguan Luo
Summary: Unmanned aerial vehicles (UAVs) have been widely applied in various fields due to their advantages of low-cost, high maneuverability, and easy operation. However, the path planning problem of UAVs remains challenging, as it directly affects flight safety and efficiency. In this study, we formulate the path planning problem as a constrained optimization problem, considering the costs of path length and threat, as well as collision and turning angle constraints. We propose a reinforcement learning-based multi-strategy cuckoo search algorithm to address the poor searchability and slow convergence speed of current optimization methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Khin Maung Htay, Rozmie Razif Othman, Amiza Amir, Jalal Mohammed Hachim Alkanaani
Summary: This paper presents a new t-way strategy based on the Gravitational Search Algorithm (GSA), known as the Gravitational Search Test Generator (GSTG), which demonstrates competitive results in most system configurations and achieves higher combination coverage.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Energy & Fuels
Imran Pervez, Charalampos Antoniadis, Yehia Massoud
Summary: Photovoltaic arrays are a simple and environmentally friendly way to generate electricity, but their power transfer efficiency depends on the electrical properties of the load, the temperature of the panels, and the sunlight conditions. Maximum Power Point Tracking (MPPT) is an optimization method that adjusts the output voltage of the PV panels to deliver maximum power to the load. However, MPPT becomes a non-convex problem when there are obstacles causing incomplete sunlight on the PV surface. Metaheuristic algorithms have better search space exploration capabilities, but there is still room for improvement in their performance.
Article
Computer Science, Theory & Methods
Weiwei Lin, Yufeng Zhang, Wentai Wu, Simon Fong, Ligang He, Jia Chang
Summary: As cloud computing technologies continue to advance, the issue of excessive energy consumption in cloud datacenters has become a widespread concern. To address this issue, we propose separate power consumption models based on workload types and an adaptive workload-aware power consumption measuring method. The method accurately predicts future workloads and selects the appropriate power model, reducing real-time power estimation lag.
Article
Computer Science, Hardware & Architecture
Jinyan Li, Yaoyang Wu, Simon Fong, Raymond K. Wong, Victor W. Chu, Kok-leong Ong, Kelvin K. L. Wong
Summary: Process mining is increasingly crucial in workflow model reconstructions, and the efficacy of this method relies on data mining algorithms accurately classifying future events from process logs. Our proposed methods address class imbalance by integrating swarm intelligence algorithms, aiming to enhance classification accuracy and confidence levels in process mining.
JOURNAL OF SUPERCOMPUTING
(2021)
Article
Computer Science, Information Systems
Ricardo Brito, Robert P. Biuk-Aghai, Simon Fong
Summary: This paper proposes a new method for improving human activity recognition datasets by adding Shadow Features to enhance the accuracy of Neural Network classifiers. NVIDIA GPU technology and the CUDA programming model are used to generate Shadow Features with little time cost, significantly improving the Neural Network accuracy.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Engineering, Civil
Yifei Tian, Wei Song, Long Chen, Simon Fong, Yunsick Sung, Jeonghoon Kwak
Summary: This paper proposes a lightweight model using a unified space autoencoder to recognize 3D objects. Experimental results show that the proposed model performs similarly to state-of-the-art models on LiDAR and ModelNet10 datasets, while having a smaller model size and shorter training time.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Computer Science, Hardware & Architecture
Jinyan Li, Yaoyang Wu, Simon Fong, Antonio J. Tallon-Ballesteros, Xin-she Yang, Sabah Mohammed, Feng Wu
Summary: This paper introduces a novel ensemble method that combines the advantages of ensemble learning and under-sampling by using a multi-objective strategy, resulting in significantly improved performance in imbalanced classification while maintaining the integrity of the original dataset. The proposed method outperforms single ensemble methods, state-of-the-art under-sampling methods, and combinations of these methods with the traditional PSO instance selection algorithm according to experimental results.
JOURNAL OF SUPERCOMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Zaid Abdi Alkareem Alyasseri, Ahamad Tajudin Khader, Mohammed Azmi Al-Betar, Xin-She Yang, Mazin Abed Mohammed, Karrar Hameed Abdulkareem, Seifedine Kadry, Imran Razzak
Summary: A multi-objective flower pollination algorithm is proposed in this study to solve the EEG signal denoising problem using wavelet transform. The algorithm optimizes the denoising parameters based on two measurement criteria, minimum mean squared error and maximum signal-to-noise ratio. Experimental results show that the proposed method achieves good performance.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Panagiotis E. Mergos, Xin-She Yang
Summary: The Flower Pollination Algorithm (FPA) is an efficient optimization algorithm inspired by the evolution process of flowering plants. In this study, a modified version of FPA called FPAPA is proposed, considering the additional feature of pollinator attraction in flower pollination. Numerical experiments show that FPAPA represents a statistically significant improvement upon the original FPA, outperforming other state-of-the-art optimization algorithms and offering better and more robust optimal solutions.
EVOLUTIONARY INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Qian Li, Sanyang Liu, Yiguang Bai, Xingshi He, Xin-She Yang
Summary: This paper investigates the robustness of complex networks under the assumption that costs are functions of node degrees. A multi-objective, elitism-based evolutionary algorithm is proposed to address the network disintegration problem. Through information retention and an update mechanism, the algorithm achieves improved convergence rate. Experimental results demonstrate that the proposed method outperforms five other state-of-the-art attack strategies.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Theory & Methods
Tengyue Li, Simon Fong, Sabah Mohammed, Jinan Fiaidhi, Steven Guan, Victor Chang
Summary: In the medical domain, data collection and model optimization for multi-class models can be time-consuming and resource-intensive. This study introduces a novel strategy to achieve maximum accuracy while significantly reducing model training time, and preliminary experiments demonstrate the feasibility of this approach.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2022)
Article
Computer Science, Hardware & Architecture
Juzhen Wang, Xiaoli Zhang, Xingshi He, Yongqiang Sun
Summary: This article investigates the scenario where multiple UAVs serve as edge computing devices for the Internet of Vehicles (IoV). By optimizing bandwidth allocation and trajectory control, the communication capacity of the system is maximized so that the UAV edge computing network can process more data. The proposed actor-critic mixing network (AC-Mix) and multi-attentive agent deep deterministic policy gradient (MA2DDPG) algorithms improve the performance compared to the benchmark algorithm MADDPG.
IEEE TRANSACTIONS ON RELIABILITY
(2023)
Article
Computer Science, Theory & Methods
Deguang You, Weiwei Lin, Fang Shi, Jianzhuo Li, Deyu Qi, Simon Fong
Summary: This paper proposes a novel prediction approach called CEEMDAN-RIDGE, which focuses on denoising to improve the accuracy of CPU load prediction by removing noise in energy consumption data. The prediction accuracy is further enhanced by error correction using historical error data. Experimental results show that the proposed approach outperforms other models in three performance metrics.
Article
Biotechnology & Applied Microbiology
Jie Yang, Jinfeng Li, Kun Lan, Anruo Wei, Han Wang, Shigao Huang, Simon Fong
Summary: There are three primary challenges in automatic diagnosis of arrhythmias: individual patient variation, complex ECG signal pathologies, and high cost of annotating clinical ECG. Traditional ECG processing relies heavily on prior knowledge, while standard deep learning methods do not fully consider the dynamic characteristics of ECG data. This paper proposes a multi-label fusion deep learning scheme for arrhythmia detection and classification, and achieves state-of-the-art performance in multi-label database experiments.
BIOENGINEERING-BASEL
(2022)
Proceedings Paper
Engineering, Electrical & Electronic
Geili. T. A. El Sanousi, Franz Hirtenfelder, Mohammed. A. H. Abbas, Raed. A. Abd-Alhameed, Xin-She Yang, Tuan Anh Le, Huan X. Nguyen
Summary: This paper introduces a novel concentric circular antenna array design with in band full duplex access and shows the effectiveness of incorporating virtual antenna formations for enhanced performance. The proposed design demonstrates excellent beam-forming abilities and IBFD reception through self-interference cancellation.
2021 28TH INTERNATIONAL CONFERENCE ON TELECOMMUNICATIONS (ICT)
(2021)
Article
Computer Science, Artificial Intelligence
J. Senthilnath, Sushant Kulkarni, S. Suresh, X. S. Yang, J. A. Benediktsson
Summary: In this study, a standalone clustering approach based on the Flower Pollination Algorithm (FPA) is proposed and demonstrated to outperform popular clustering algorithms and metaheuristic algorithms.
EVOLUTIONARY INTELLIGENCE
(2021)