Article
Computer Science, Artificial Intelligence
Prachi Agrawal, Talari Ganesh, Ali Wagdy Mohamed
Summary: The study proposed two approaches to solve feature selection problems: FS-BGSK and FS-pBGSK, which showed better accuracy, convergence, and robustness in experiments.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Engineering, Multidisciplinary
Esin Ayse Zaimoglu, Nilufer Yurtay, Huseyin Demirci, Yuksel Yurtay
Summary: Feature selection is a challenging and common problem in machine learning. This paper proposes a binary chaotic horse herd optimization algorithm for feature selection (BCHOAFS). Experimental results show that the proposed method outperforms or competes with other methods in terms of accuracy.
ENGINEERING SCIENCE AND TECHNOLOGY-AN INTERNATIONAL JOURNAL-JESTECH
(2023)
Article
Mathematics, Applied
Cheng Luo, Bao-Qing Liu, Hu-Shuang Hou
Summary: In this study, q-deformation of fractional chaotic maps was investigated by revisiting deformation results in difference equations or chaotic maps. Fractional differences and q-deformations were introduced, leading to the proposal of new fractional chaotic maps with a q-parameter. Chaotic behaviors were discussed in both one and two dimensional cases, and stability analysis of generalized Henon maps was provided with numerical results.
APPLIED MATHEMATICS AND COMPUTATION
(2021)
Article
Engineering, Mechanical
Mark Edelman
Summary: The paper discusses the significance of power-law memory in natural and social systems, as well as the use of fractional calculus to describe system behavior. The behavior of fractional systems differs from systems without memory, and finding periodic points is crucial. Fractional systems do not have periodic points but only asymptotically periodic sinks.
NONLINEAR DYNAMICS
(2021)
Article
Computer Science, Interdisciplinary Applications
Guojiang Xiong, Xufeng Yuan, Ali Wagdy Mohamed, Jun Chen, Jing Zhang
Summary: Fault section location plays a critical role in power distribution networks. This study proposes an improved binary gaining-sharing knowledge-based algorithm (IBGSK) to effectively solve the fault section location task. IBGSK outperforms other advanced algorithms in solution quality, robustness, and speed, showing promising potential for distribution networks.
JOURNAL OF COMPUTATIONAL DESIGN AND ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Utkarsh Agrawal, Vasudha Rohatgi, Rahul Katarya
Summary: The problem of feature selection involves selecting the most informative subset of features that have the most impact in classification. This paper proposes a novel variant of the Equilibrium Optimizer called Normalized Mutual Information-based equilibrium optimizer (NMIEO) for feature selection. The proposed method incorporates a local search strategy based on Normalized Mutual Information and utilizes Chaotic maps for population initialization. Experimental results demonstrate the superior performance of NMIEO compared to other competitive methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Laith Abualigah, Ali Diabat
Summary: This paper proposes a feature selection method called CGSO that combines chaotic maps and binary Group Search Optimizer. Experimental results demonstrate the superiority of this method over other published methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Prachi Agrawal, Talari Ganesh, Ali Wagdy Mohamed
Summary: This article introduces a novel binary algorithm NBGSK to solve binary optimization problems, utilizing two binary stages and a knowledge factor to explore and exploit the search space efficiently, while also introducing PR-NBGSK to enhance performance and prevent trapping into local optima. Applied to knapsack instances, NBGSK and PR-NBGSK show better efficiency and effectiveness in terms of convergence, robustness, and accuracy.
COMPLEX & INTELLIGENT SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Guojiang Xiong, Xufeng Yuan, Ali Wagdy Mohamed, Jing Zhang
Summary: LOBGSK is an improved binary variant of knowledge-sharing algorithm designed for fault section diagnosis in power systems. By utilizing binary encoding and logical operations, it can quickly and accurately diagnose various faults with a high success rate, outperforming other metaheuristic algorithms.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Benedict Jun Ma, Shuai Liu, Ali Asghar Heidari
Summary: This paper proposes a hunger games search algorithm integrated with a multi-strategy framework for feature selection. Through comparative experiments on benchmark functions and datasets, the algorithm demonstrates superior performance in terms of classification accuracy and the number of selected features.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Yuxin Zhao, Junwei Dong, Xiaobo Li, Hui Chen, Shaolang Li
Summary: Feature selection is important in data mining and pattern recognition, as it reduces computation time and improves classification performance. This paper proposes a binary dandelion algorithm (BDA) and an improved version called SBDA, which incorporates a seeding strategy and chaotic populations. Experimental results demonstrate that SBDA achieves higher classification accuracy in most cases.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Ziwei Li, Sai Ravela
Summary: This study examines the use of artificial neural networks (NNs) to model chaotic dynamics. It shows from a geometric perspective that NNs can efficiently emulate chaotic dynamics by becoming structurally chaotic themselves. The research confirms the effectiveness of NNs in reconstructing strange attractors, extrapolating outside training data boundaries, and accurately predicting local divergence rates. The authors propose that the trained network's map involves sequential geometric stretching, rotation, and compression operations, indicating topological mixing and chaos, which explains why NNs are naturally suitable for emulating chaotic dynamics.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Physics, Multidisciplinary
Maide Bucolo, Arturo Buscarino, Luigi Fortuna, Salvina Gagliano
Summary: This paper presents the general concept of multidimensional discrete maps and shows the invariance of bifurcation points from periodic to chaotic behavior. Numerical examples of the logistic map, complex-valued Ikeda map, and multivariable Henon map in multidimensional cases are reported.
FRONTIERS IN PHYSICS
(2022)
Article
Materials Science, Multidisciplinary
Shaofu Wang
Summary: By employing linear entropy, a discrete quantum logistic chaotic map is introduced to study the dynamical evolution of the quantum radiation field interacting with matter by changing independent parameters. The effects observed include chaotic properties of quantum optics and a quantum classical correspondence. Coherence dynamics of phase space is used to reveal the chaos and regular structure. Furthermore, an approach for fast adaptive synchronization of the discrete quantum maps is proposed, with results validating its effectiveness.
RESULTS IN PHYSICS
(2023)
Article
Computer Science, Artificial Intelligence
Bita Hallajian, Homayun Motameni, Ebrahim Akbari
Summary: This paper proposes a novel feature selection method that improves classification accuracy by removing irrelevant and redundant features. The method utilizes a distance-based relevancy-redundancy measurement and combines supervised and unsupervised methods to select the most relevant subset of features. Experimental results demonstrate the superiority of this method in terms of stability, classification accuracy, and other performance metrics compared to existing methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Ali Wagdy Mohamed, Karam M. Sallam, Prachi Agrawal, Anas A. Hadi, Ali Khater Mohamed
Summary: This paper evaluates the performance of various developed meta-heuristic algorithms on the recently developed CEC 2021 benchmark functions. Based on the experimental results, observations, recommendations, and conclusions are provided.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Thermodynamics
Zaiyu Gu, Guojiang Xiong, Xiaofan Fu, Ali Wagdy Mohamed, Mohammed Azmi Al-Betar, Hao Chen, Jun Chen
Summary: Photovoltaic power generation is crucial for environmental protection and requires accurate modeling and parameter extraction. The proposed ELADE algorithm combines multiple strategies to improve the performance of differential evolution in achieving accurate parameters for photovoltaic cell models. Experimental results show that a population size of 50 produces the most reliable parameters compared to other algorithms. Statistical tests confirm the superiority of ELADE, with the parameters adaptive strategy being the most influential.
ENERGY CONVERSION AND MANAGEMENT
(2023)
Article
Computer Science, Artificial Intelligence
Guojiang Xiong, Lei Li, Ali Wagdy Mohamed, Jing Zhang, Yao Zhang, Hao Chen
Summary: Establishing an accurate equivalent model is crucial for studying and analyzing the energy conversion characteristics of photovoltaic systems. However, existing equivalent models are highly nonlinear and have unknown parameters, making parameter identification difficult. To address this issue, a dual-population gaining-sharing knowledge-based algorithm (DPGSK) is proposed, which introduces a dual-population evolution strategy to improve searchability. DPGSK achieves accurate and reliable results, demonstrating its superiority over other algorithms in solving this problem.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2023)
Article
Mathematics
Akash Saxena, Ramadan A. Zeineldin, Ali Wagdy Mohamed
Summary: Energy is crucial for the development of a country, and its consumption, production, and transition to green energy are essential for sustainable development. Forecasting technologies, especially grey systems, are gaining attention due to their ability to analyze a limited amount of data. In this study, an optimized grey machine learning model using a polynomial structure was used to predict power generation, consumption, and CO2 emissions, outperforming conventional grey models in terms of accuracy.
Article
Computer Science, Artificial Intelligence
Samah Mohamed, Hazem A. A. Nomer, Retaj Yousri, Ali Wagdy Mohamed, Ahmed Soltan, M. Saeed Darweesh
Summary: Wearable devices, including wearable medical devices, are a growing field of research with various applications. Power saving is crucial for such devices due to their limited power supply. This paper proposes a task scheduler for wearable medical devices based on a Gaining-Sharing Knowledge (GSK) algorithm to optimize energy consumption. The effectiveness of the GSK-based scheduling algorithm is evaluated against existing techniques using experimental data collected from a prototype.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Computer Science, Interdisciplinary Applications
Qinghua Liu, Guojiang Xiong, Xiaofan Fu, Ali Wagdy Mohamed, Jing Zhang, Mohammed Azmi Al-Betar, Hao Chen, Jun Chen, Sheng Xu
Summary: This study proposes a new hybrid method, GSK-DE, to solve large-scale ED problems by integrating the advantages of GSK and DE algorithms. By dividing the population into two subpopulations, one performing GSK and the other executing DE, and combining the updated individuals, GSK-DE improves the searching efficiency. Simulation results demonstrate that GSK-DE achieves quicker global convergence, higher quality dispatch schemes, and greater robustness.
JOURNAL OF COMPUTATIONAL DESIGN AND ENGINEERING
(2023)
Article
Mathematics, Applied
Golla Madhu, Sandeep Kautish, Khalid Abdulaziz Alnowibet, Hossam M. M. Zawbaa, Ali Wagdy Mohamed
Summary: In recent years, various deep neural networks have been widely used in applications such as medical diagnosis, image analysis, and self-driving vehicles. The choice of activation function in deep neural networks greatly impacts the training and reliability of the models. While Rectified Linear Unit (ReLU) has been the most popular activation function, it has some flaws like dying ReLU and bias shift. This research proposes a new activation function called NIPUNA, which outperforms traditional activation functions when tested on customized convolutional neural networks (CCNN) trained on benchmark datasets like Fashion MNIST and MNIST.
Review
Social Sciences, Interdisciplinary
Muzaffar Hamzah, Md. Monirul Islam, Shahriar Hassan, Md. Nasim Akhtar, Most. Jannatul Ferdous, Muhammed Basheer Jasser, Ali Wagdy Mohamed
Summary: Cyber-Physical System (CPS) is a symbol of the fourth industrial revolution (4IR) by integrating physical and computational processes, which can associate with humans in various ways. It is assisting to incorporate the world and influencing our ordinary life significantly.
Article
Computer Science, Theory & Methods
Ahmed M. Anter, Ali W. Mohamed, Min Zhang, Zhiguo Zhang
Summary: Parkinson's disease (PD) is a degenerative neurological disease, and early diagnosis is crucial. Monitoring PD progression from voice records is a promising technique for IoT-based telemedicine in smart homes. However, selecting the most relevant voice features fast and accurately for early diagnosis is still an open question.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2023)
Article
Engineering, Chemical
Shoyab Ali, Annapurna Bhargava, Akash Saxena, Abdulaziz S. Almazyad, Karam M. Sallam, Ali Wagdy Mohamed
Summary: Hybrid Active Power Filter (HAPF) is an important technology for mitigating harmonic pollution in electrical systems. This paper proposes an estimator that accurately estimates the parameters of HAPF configuration using an Amended Crow Search Algorithm (ACSA). The results show that the proposed algorithm achieves optimal results in reducing harmonic pollution.
Article
Engineering, Multidisciplinary
Mohamed Abdel-Basset, Reda Mohamed, Muhammed Basheer Jasser, Ibrahim M. Hezam, Karam M. Sallam, Ali Wagdy Mohamed
Summary: This paper presents a new variant of the artificial gorilla troops optimizer (GTO) called ranking-based GTO (RGTO), which uses two strategies to improve its exploitation and exploration capabilities. The algorithm is evaluated using a benchmark and demonstrates outstanding performance for three engineering optimization problems.
ALEXANDRIA ENGINEERING JOURNAL
(2023)
Article
Engineering, Multidisciplinary
Vanita Garg, Kusum Deep, Khalid Abdulaziz Alnowibet, Hossam M. Zawbaa, Ali Wagdy Mohamed
Summary: In this paper, a novel attempt is made to combine two effective algorithm strategies, with BBO focusing on exploration and SSA focusing on exploitation. The proposed algorithm is evaluated using IEEE CEC 2014 and statistical convergence graphs are provided. Additionally, the algorithm is applied to 10 real life problems and compared with its counterpart algorithm, demonstrating the superior performance of the hybrid version of BBO.
ALEXANDRIA ENGINEERING JOURNAL
(2023)
Article
Mathematics
Nour Elhouda Chalabi, Abdelouahab Attia, Khalid Abdulaziz Alnowibet, Hossam M. Zawbaa, Hatem Masri, Ali Wagdy Mohamed
Summary: This paper proposes an extended version of the gaining-sharing knowledge optimization (GSK) algorithm, named multiobjective gaining-sharing knowledge optimization (MOGSK), to deal with multiobjective optimization problems. The MOGSK algorithm employs an external archive population to guide the solutions during the exploration process and incorporates fast nondominated sorting with crowding distance to ensure diversity and convergence. Experimental results demonstrate the effectiveness of the proposed MOGSK algorithm in real-world optimization problems.
Article
Computer Science, Information Systems
Narayanan Ganesh, Sambandan Jayalakshmi, Rama Chandran Narayanan, Miroslav Mahdal, Hossam M. M. Zawbaa, Ali Wagdy Mohamed
Summary: Image classification, an important aspect of image processing, is challenging and time-consuming. Traditional models have limitations in effectiveness and manual procedures become inefficient due to the large volume of data. In this study, a deep learning-based classification model is developed to improve accuracy and handle large datasets. Various techniques, such as the Adaptive Guided Bilateral Filter and Spectral Gabor Wavelet Transform, are used for image filtering and feature extraction. The proposed model achieved an accuracy of 98.8% when tested on a brain tumor MRI dataset in the MATLAB platform.
Article
Computer Science, Information Systems
A. Reyana, Sandeep Kautish, P. M. Sharan Karthik, Ibrahim Ahmed Al-Baltah, Muhammed Basheer Jasser, Ali Wagdy Mohamed
Summary: Farmers and agronomists are using sensors and IoT to remotely monitor crops and improve agriculture operations. This paper presents a novel Multisensor Machine-Learning Approach (MMLA) for classifying multisensor data, providing cultivation recommendations and increasing crop yield. The proposed recommendation system classifies eight crop species using machine learning algorithms and shows promising results compared to state-of-the-art classifiers.