Article
Computer Science, Interdisciplinary Applications
Jing Cao, Juncheng Gao, Hima Nikafshan Rad, Ahmed Salih Mohammed, Mahdi Hasanipanah, Jian Zhou
Summary: The study aims to propose an efficient machine learning model to predict engineering properties of rock, with the XGBoost-FA model showing superior accuracy and generalization compared to other models.
ENGINEERING WITH COMPUTERS
(2022)
Article
Construction & Building Technology
Hongyan Wang, Wen Wen, Zihong Zhang, Ning Gao
Summary: This study uses the optimized Relevance Vector Machine (RVM) model with Sparrow Search Algorithm (SSA), Simulate Anneal Arithmetic (SAA), Particle Swarm Optimization (PSO), and Bayesian Optimization Algorithm (BOP) to construct an energy dissipation model for public buildings in Wuhan City. The study finds that building area, personnel density, and supply air temperature significantly impact energy dissipation in public buildings. By employing Principal Component Analysis (PCA) for dimensionality reduction, the study selects seven main influential factors to predict building energy consumption accurately. The BOP-RVM model performs well in terms of R-2 (0.9523), r (0.9761), RMSE (5.3894), and SI (0.056), providing practical value for energy management strategies.
Article
Computer Science, Artificial Intelligence
Miodrag Zivkovic, Milan Tair, K. Venkatachalam, Nebojsa Bacanin, Stepan Hubalovsky, Pavel Trojovsky
Summary: This article presents an improved firefly algorithm for optimizing XGBoost classifier hyperparameters in order to improve the accuracy of network intrusion detection systems. Experimental results demonstrate the significant potential of the proposed algorithm in machine learning hyperparameter optimization.
PEERJ COMPUTER SCIENCE
(2022)
Article
Computer Science, Artificial Intelligence
M. Revanesh, S. A. Sahaaya Arul Mary, G. Gnaneswari, G. Maria Jones, K. V. Kanimozhi, G. K. Kamalam
Summary: The wireless sensor nodes collect environmental data while using their batteries to communicate. Sharing data consumes more energy and shortens the network lifetime. Energy efficiency is crucial, and the proposed technique optimizes cluster head selection to enhance deep neural network performance. This technique improves search activity and increases network longevity. Throughput, leftover energy, and active nodes are evaluated, and the proposed approach outperforms the firefly algorithm.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Multidisciplinary Sciences
Bilel Zerouali, Celso Augusto Guimaraes Santos, Camilo Allyson Simoes de Farias, Raul Souza Muniz, Salah Difi, Zaki Abda, Mohamed Chettih, Salim Heddam, Samy A. Anwar, Ahmed Elbeltagi
Summary: This study proposes a multilayer perceptron (MLP) network optimized by Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Firefly Algorithm (FFA), and Tel-econnection Pattern Indices to model monthly rainfalls at the Sebaou River basin. The NAO index was found to be the most influential parameter in improving the modeling accuracy. The comparison between the best MLP model and Bat-ELM revealed the high performance of Bat-ELM for rainfall modeling at the Sebaou River basin.
Article
Chemistry, Physical
Jiandong Huang, Mengmeng Zhou, Hongwei Yuan, Mohanad Muayad Sabri Sabri, Xiang Li
Summary: In this study, a hybrid machine learning model combining Random Forests (RF) and Firefly Algorithm (FA) was proposed to predict the compressive strength of metakaolin cement-based materials. The importance of cement grade and water-to-binder ratio on compressive strength was found to be the highest.
Article
Engineering, Environmental
Yuantian Sun, Guichen Li, Junfei Zhang, Jiandong Huang
Summary: The study proposed an ensemble classifier RF-FA model for rockburst prediction, which effectively optimized the hyperparameters of RF using FA. By selecting key parameters as input variables and rockburst intensity as output, the model demonstrated high performance in independent test sets and new engineering projects, showing better accuracy compared to existing models.
BULLETIN OF ENGINEERING GEOLOGY AND THE ENVIRONMENT
(2021)
Article
Chemistry, Multidisciplinary
Lei Qiao, Zhining Jia, You Cui, Kun Xiao, Haonan Su
Summary: This study proposes a method based on the improved sparrow search algorithm optimized deep extreme learning machine (ISSA-DELM) to enhance the accuracy and stability of DTS prediction. Experimental results show that the comprehensive performance of this method is better than other approaches, providing an effective solution for estimating missing DTS values.
APPLIED SCIENCES-BASEL
(2022)
Article
Environmental Sciences
Naser Arya Azar, Zahra Kayhomayoon, Sami Ghordoyee Milan, HamedReza Zarif Sanayei, Ronny Berndtsson, Zahra Nematollahi
Summary: Due to limited groundwater resources, the conjunctive use of surface and groundwater is important. Researchers developed a comprehensive methodological structure using numerical models, optimization algorithms, and machine learning to optimize the allocation and extraction of water resources. They tested the methodology in an important aquifer in Iran and found that machine learning models are cost- and time-effective solutions for estimating optimal exploitation of groundwater resources.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2022)
Article
Agronomy
Lifeng Wu, Youwen Peng, Junliang Fan, Yicheng Wang, Guomin Huang
Summary: A novel machine learning model, Kmeans-FFA-KELM, was proposed for estimating monthly mean daily ET0, which outperformed other models in terms of prediction accuracy and significantly reduced computational time.
AGRICULTURAL WATER MANAGEMENT
(2021)
Article
Green & Sustainable Science & Technology
Yong Liu, Yang Wang, Mengmeng Zhou, Jiandong Huang
Summary: It is important to mix discarded concrete blocks with gradation and use them as aggregate for recycled concrete to promote sustainable pavement development. The traditional evaluation model for compressive strength of recycled concrete is not efficient, thus the firefly algorithm is proposed to optimize the random forest model. The results show that the hybrid model significantly improves computational efficiency and accuracy.
Article
Computer Science, Artificial Intelligence
Janmenjoy Nayak, Bighnaraj Naik, Pandit Byomakesha Dash, Alireza Souri, Vimal Shanmuganathan
Summary: This paper proposes an efficient hand gesture recognition method based on LightGBM and memetic firefly algorithm, achieving a high accuracy of 99.36% and robust reliability.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Information Systems
Hu Peng, Wenhua Zhu, Changshou Deng, Zhijian Wu
Summary: The firefly algorithm (FA) is a nature-inspired heuristic optimization algorithm based on the luminescence and attraction behavior of fireflies. A novel courtship learning (CL) framework is proposed to enhance the performance of the FA by dividing the population into female and male subpopulations. Experimental results confirm that the proposed CL framework significantly enhances the performance of the original FA and advanced FA variants.
INFORMATION SCIENCES
(2021)
Article
Engineering, Multidisciplinary
Nadir Omer, Ahmed H. Samak, Ahmed I. Taloba, Rasha M. Abd El-Aziz
Summary: Nowadays, the Internet is widely used for daily necessities, leading to an increase in cyberattacks and cybercrime. Machine learning techniques are being utilized to detect network attacks and enhance cybersecurity. Developing intrusion detection systems can identify anomalies and improve overall security, and an efficient system will be created using machine learning techniques.
ALEXANDRIA ENGINEERING JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Asma M. Altabeeb, Abdulqader M. Mohsen, Laith Abualigah, Abdullatif Ghallab
Summary: The study introduces a cooperative hybrid firefly algorithm to solve the capacitated vehicle routing problem (CVRP), which utilizes multiple firefly algorithm populations to collaborate, hybridizes with local search and genetic operators, and exchanges solutions among populations through communication, the results of experiments demonstrate the algorithm's outstanding performance compared to other methods.
APPLIED SOFT COMPUTING
(2021)
Article
Environmental Sciences
Marzieh Fadaee, Amin Mahdavi-Meymand, Mohammad Zounemat-Kermani
Summary: This study investigates the application of two metaheuristic algorithms (BOA and GA) in the prediction of Suspended Sediment Load (SSL). The results show that the performances of all models are similar, and the metaheuristic algorithms can improve the accuracy of some models, with BOA outperforming GA.
GEOCARTO INTERNATIONAL
(2022)
Article
Engineering, Civil
Amin Mahdavi-Meymand, Wojciech Sulisz
Summary: This study integrates multiple optimization algorithms with support vector regression to predict energy dissipation downstream of labyrinth weirs. The results indicate that meta-heuristic algorithms greatly improve the performance of support vector regression, with the SVR-MTOA method yielding the best results.
JOURNAL OF HYDRO-ENVIRONMENT RESEARCH
(2022)
Article
Engineering, Multidisciplinary
Barbara Stachurska, Amin Mahdavi-Meymand, Wojciech Sulisz
Summary: The study developed new integrated machine learning methods using the SPBO algorithm to determine wave-induced non-cohesive sediment particle velocities. The techniques successfully predict the sediment particle horizontal velocity using easily measurable data, showing potential for interpreting measurement data and verifying sediment transport models.
Article
Multidisciplinary Sciences
Amin Mahdavi-Meymand, Mohammad Zounemat-Kermani, Wojciech Sulisz, Rodolfo Silva
Summary: This study develops machine learning models to predict wave run-up height and enhances the accuracy by employing optimization algorithms. The results indicate that these ML models are more accurate than empirical relations, with the GMDH-FA and GMDH-IWO models being recommended for applications in coastal engineering.
SCIENTIFIC REPORTS
(2022)
Article
Engineering, Marine
Amin Mahdavi-Meymand, Mohammad Zounemat-Kermani
Summary: In this study, the Mayfly Algorithm (MA) was applied to predict the depth and diameter of scour holes caused by hydro-suction. Machine learning methods based on MA and Genetic Algorithm (GA) were trained, and the results showed that MA improved the accuracy.
SHIPS AND OFFSHORE STRUCTURES
(2022)
Article
Computer Science, Artificial Intelligence
Amin Mahdavi-Meymand, Mohammad Zounemat-Kermani
Summary: This study introduces a novel meta-heuristic algorithm called Homonuclear Molecules Optimization (HMO) for optimizing complex and nonlinear problems. HMO is inspired by the arrangement of electrons around atoms and the structure of homonuclear molecules. The algorithm effectively solves various optimization problems and outperforms classical and novel algorithms.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Multidisciplinary Sciences
Fahime Javadi, Kourosh Qaderi, Mohammad Mehdi Ahmadi, Majid Rahimpour, Mohamad Reza Madadi, Amin Mandavi-Meymand
Summary: This study investigated the capabilities of machine learning models in predicting sediment discharge in free-flow flushing. The results showed that the HGSO and EO algorithms improved the accuracy of the GMDH model, and SVR-EO and SVR-HGSO provided the highest accuracy in both training and validation phases, with GMDH-HGSO being the most accurate model.
SCIENTIFIC REPORTS
(2022)
Article
Green & Sustainable Science & Technology
Amin Mahdavi-Meymand, Wojciech Sulisz
Summary: In this study, nested artificial neural networks were developed and applied to predict significant wave height at twenty selected locations of the North Sea, using wind speed and wind direction as input parameters. The results showed that the derived models were 18.39% more accurate than linear regression, and the nested artificial neural network could increase the accuracy of traditional models by up to 34%. Among all applied models, the nested artificial neural network developed based on the integration of particle swarm optimization algorithm and adaptive neuro-fuzzy inference system provided the most accurate prediction of wave heights, with RMSE = 0.525m and R2 = 0.84. The high accuracy of the results suggests that the application of nested artificial neural networks may be recommended for modeling wave parameters and other complex problems, if computational time is not critical for users.
Article
Environmental Sciences
Ozgur Kisi, Kulwinder Singh Parmar, Amin Mahdavi-Meymand, Rana Muhammad Adnan, Shamsuddin Shahid, Mohammad Zounemat-Kermani
Summary: This study investigated the potential of four different neuro-fuzzy embedded meta-heuristic algorithms (particle swarm optimization, genetic algorithm, harmony search, and teaching-learning-based optimization algorithm) in estimating the water quality of the Yamuna River in Delhi, India. The performance of hybrid neuro-fuzzy models in predicting COD was compared with classical neuro-fuzzy and LSSVM methods, with higher accuracy achieved when certain water quality parameters were used as inputs. The results demonstrated that the neuro-fuzzy models optimized with harmony search provided the best accuracy, while the particle swarm optimization and teaching-learning-based optimization showed the highest computational speed.
Article
Geosciences, Multidisciplinary
Amin Mahdavi-Meymand, Wojciech Sulisz
Summary: An original particle swarm clustered optimization (PSCO) method was developed for implementation in applied sciences, showing better performance than traditional particle swarm optimization algorithms. PSCO was integrated with various techniques to predict river discharge, and the results indicated its accuracy and improvement in machine learning techniques. The ANFIS-PSCO model with RMSE = 108.433 and R-2 = 0.961 was found to be the most accurate.
PROGRESS IN EARTH AND PLANETARY SCIENCE
(2023)
Review
Computer Science, Interdisciplinary Applications
Amin Mahdavi-Meymand, Wojciech Sulisz, Mohammad Zounemat-Kermani
Summary: It has been claimed that the combination of meta-heuristic algorithms (MHAs) with base learners can improve the accuracy of hydrological machine learning (ML) models. However, a drawback of this approach is the high computing cost. Among the investigated MHAs, genetic algorithm (GA) is widely used but particle swarm optimization algorithm produces more accurate results and is recommended for hydrological applications.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2023)
Article
Engineering, Marine
Amin Mahdavi-Meymand, Wojciech Sulisz
Summary: This study developed novel clustered machine learning (CML) models and compared their performance with regular machine learning algorithms. The results showed that the clustered models outperformed the regular models, demonstrating better predictive accuracy.
Article
Environmental Sciences
Amin Mahdavi-Meymand, Wojciech Sulisz
Summary: Novel data-driven models were developed to predict water surface levels for two stations of the Vistula river. These models, trained using particle swarm optimization (PSO) and teaching-learning-based optimization (TLBO), outperformed regression equations for both large and small datasets. The N-GMDH model yielded higher accuracy than the standard GMDH, with a difference of 4.68%. TLBO also showed higher accuracy (2.06%) compared to PSO and demonstrated greater stability in finding global solutions.
INTERNATIONAL JOURNAL OF HYDROLOGY SCIENCE AND TECHNOLOGY
(2023)
Article
Engineering, Chemical
Amin Mahdavi-Meymand, Wojciech Sulisz
Summary: In this study, ARIA models were developed to enhance the prediction of boiling point rise in a multi-stage flash desalination system. The ARIA models showed greater accuracy and increased prediction efficiency compared to regular models. The ARIA-ANFIS model performed the best, reducing the error in RF predictions by 69.66%.