Search

Feature selection is becoming a relevant problem within the field of machine learning. The feature selection problem focuses on the selection of the small, necessary, and sufficient subset of features that represent the general set of features, eliminating redundant and irrelevant information. Given the importance of the topic, in recent years there has been a boom in the study of the problem, generating a large number of related investigations. Given this, this work analyzes 161 articles published between 2019 and 2023 (20 April 2023), emphasizing the formulation of the problem and performance measures, and proposing classifications for the objective functions and evaluation metrics. Furthermore, an in-depth description and analysis of metaheuristics, benchmark datasets, and practical real-world applications are presented. Finally, in light of recent advances, this review paper provides future research opportunities. [ABSTRACT FROM AUTHOR]

The efficiency and the effectiveness of a machine learning (ML) model are greatly influenced by feature selection (FS), a crucial preprocessing step in machine learning that seeks out the ideal set of characteristics with the maximum accuracy possible. Due to their dominance over traditional optimization techniques, researchers are concentrating on a variety of metaheuristic (or evolutionary) algorithms and trying to suggest cutting-edge hybrid techniques to handle FS issues. The use of hybrid metaheuristic approaches for FS has thus been the subject of numerous research works. The purpose of this paper is to critically assess the existing hybrid FS approaches and to give a thorough literature review on the hybridization of different metaheuristic/evolutionary strategies that have been employed for supporting FS. This article reviews pertinent documents on hybrid frameworks that were published in the period from 2009 to 2022 and offers a thorough analysis of the used techniques, classifiers, datasets, applications, assessment metrics, and schemes of hybridization. Additionally, new open research issues and challenges are identified to pinpoint the areas that have to be further explored for additional study. [ABSTRACT FROM AUTHOR]

Feature Selection (FS) is an essential process that is implicated in data mining and machine learning for data preparation by removing redundant and irrelevant features, thereby falling the possible risk associated with the expletive of dimensionality caused by the large dataset. As a result, FS is thought to be a combinatorial NP-hard problem, which refers to a situation where the computation time increases as the problem dimension increases. Recently, researchers have focused on metaheuristic algorithms to perform this task. Therefore, this paper proposes an effective metaheuristic which is a new variant of the recently reported Golden Jackel Optimization (GJO) algorithm called Improved GJO (IGJO). The basic GJO algorithm suffers from a local optima trap when handling large dimensional feature selection problems. Therefore, the effectiveness of the GJO is improved by considering the operators from the gradient-based optimizer. The proposed IGJO is based on the local escaping operator and the direction of population movement to improve the exploration and exploitation ability of the basic GJO algorithm. The superiority of the IGJO algorithm is tested on 23 standard numerical benchmark problems, 29 CEC2017 optimization problems, and 33 CEC2020 constrained real-world engineering design problems. Additionally, the IGJO is transformed to its binary version for the FS problem using a new nonlinear time-varying sigmoid transfer function, and finally, the binary variant is validated on FS problems with different benchmark datasets. The performance of the IGJO is compared with well-known algorithms to validate its superiority. The obtained results show that the IGJO is a reliable tool for numerical optimization problems and FS problems. [ABSTRACT FROM AUTHOR]

The hybrid genetic search (HGS) metaheuristic has produced outstanding results for several variants of the vehicle routing problem. A recent implementation of HGS specialized to the capacitated vehicle routing problem (CVRP) is a state‐of‐the‐art method for this variant. This paper proposes an improved HGS for the CVRP obtained by incorporating a new solution generation method into its (re‐)initialization process to guide the search more efficiently and effectively. The solution generation method introduced in this work combines an approach based on frequent patterns extracted from good solutions by a data mining process and a randomized version of the Clarke and Wright savings heuristic. As observed in our experimental comparison, the proposed method significantly outperforms the original algorithm regarding the final gap to the best known solutions and the primal integral. [ABSTRACT FROM AUTHOR]

In recent years, metaheuristic methods have shown remarkable efficacy in resolving complex combinatorial challenges across a broad spectrum of fields. Nevertheless, the escalating complexity of these problems necessitates the continuous development of innovative techniques to enhance the performance and reliability of these methods. This paper aims to contribute to this endeavor by examining the impact of solution initialization methods on the performance of a hybrid algorithm applied to the set union knapsack problem (SUKP). Three distinct solution initialization methods, random, greedy, and weighted, have been proposed and evaluated. These have been integrated within a sine cosine algorithm employing k-means as a binarization procedure. Through testing on medium- and large-sized SUKP instances, the study reveals that the solution initialization strategy influences the algorithm's performance, with the weighted method consistently outperforming the other two. Additionally, the obtained results were benchmarked against various metaheuristics that have previously solved SUKP, showing favorable performance in this comparison. [ABSTRACT FROM AUTHOR]

Intrusion detection systems (IDS) play a crucial role in securing networks and identifying malicious activity. This is a critical problem in cyber security. In recent years, metaheuristic optimization algorithms and deep learning techniques have been applied to IDS to improve their accuracy and efficiency. Generally, optimization algorithms can be used to boost the performance of IDS models. Deep learning methods, such as convolutional neural networks, have also been used to improve the ability of IDS to detect and classify intrusions. In this paper, we propose a new IDS model based on the combination of deep learning and optimization methods. First, a feature extraction method based on CNNs is developed. Then, a new feature selection method is used based on a modified version of Growth Optimizer (GO), called MGO. We use the Whale Optimization Algorithm (WOA) to boost the search process of the GO. Extensive evaluation and comparisons have been conducted to assess the quality of the suggested method using public datasets of cloud and Internet of Things (IoT) environments. The applied techniques have shown promising results in identifying previously unknown attacks with high accuracy rates. The MGO performed better than several previous methods in all experimental comparisons. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) technologies has gained significant interest in the design of smart grids (SGs). The increasing amount of distributed generations, maturity of existing grid infrastructures, and demand network transformation have received maximum attention. An essential energy storing model mostly the electrical energy stored methods are developing as the diagnoses for its procedure was becoming further compelling. The dynamic electrical energy stored model using Electric Vehicles (EVs) is comparatively standard because of its excellent electrical property and flexibility however the chance of damage to its battery was there in event of overcharging or deep discharging and its mass penetration deeply influences the grids. This paper offers a new Hybridization of Bacterial foraging optimization with Sparse Autoencoder (HBFOA-SAE) model for IoT Enabled energy systems. The proposed HBFOA-SAE model majorly intends to effectually estimate the state of charge (SOC) values in the IoT based energy system. To accomplish this, the SAE technique was executed to proper determination of the SOC values in the energy systems. Next, for improving the performance of the SOC estimation process, the HBFOA is employed. In addition, the HBFOA technique is derived by the integration of the hill climbing (HC) concepts with the BFOA to improve the overall efficiency. For ensuring better outcomes for the HBFOA-SAE model, a comprehensive set of simulations were performed and the outcomes are inspected under several aspects. The experimental results reported the supremacy of the HBFOA-SAE model over the recent state of art approaches. [ABSTRACT FROM AUTHOR]

As the volume of data generated by information systems continues to increase, machine learning (ML) techniques have become essential for the extraction of meaningful insights. However, the sheer volume of data often causes these techniques to become sluggish. To overcome this, feature selection is a vital step in the pre-processing of data. In this paper, we introduce a novel K-nearest neighborhood (KNN)-based wrapper system for feature selection that leverages the iterative improvement ability of the weighted superposition attraction (WSA). We evaluate the performance of WSA against seven well-known metaheuristic algorithms, i.e., differential evolution (DE), genetic algorithm (GA), particle swarm optimization (PSO), flower pollination algorithm (FPA), symbiotic organisms search (SOS), marine predators' algorithm (MPA) and manta ray foraging optimization (MRFO). Our extensive numerical experiments demonstrate that WSA is highly effective for feature selection, achieving a decrease of up to 99% in the number of features for large datasets without sacrificing classification accuracy. In fact, WSA-KNN outperforms traditional ML methods by about 18% and ensemble ML algorithms by 9%. Moreover, WSA-KNN achieves comparable or slightly better solutions when compared with neural networks hybridized with metaheuristics. These findings highlight the importance and potential of WSA for feature selection in modern-day data processing systems. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) is a modern approach that enables connection with a wide variety of devices remotely. Due to the resource constraints and open nature of IoT nodes, the routing protocol for low power and lossy (RPL) networks may be vulnerable to several routing attacks. That's why a network intrusion detection system (NIDS) is needed to guard against routing assaults on RPL-based IoT networks. The imbalance between the false and valid attacks in the training set degrades the performance of machine learning employed to detect network attacks. Therefore, we propose in this paper a novel approach to balance the dataset classes based on metaheuristic optimization applied to locality-sensitive hashing and synthetic minority oversampling technique (LSH-SMOTE). The proposed optimization approach is based on a new hybrid between the grey wolf and dipper throated optimization algorithms. To prove the effectiveness of the proposed approach, a set of experiments were conducted to evaluate the performance of NIDS for three cases, namely, detection without dataset balancing, detection with SMOTE balancing, and detection with the proposed optimized LSHSOMTE balancing. Experimental results showed that the proposed approach outperforms the other approaches and could boost the detection accuracy. In addition, a statistical analysis is performed to study the significance and stability of the proposed approach. The conducted experiments include seven different types of attack cases in the RPL-NIDS17 dataset. Based on the proposed approach, the achieved accuracy is (98.1%), sensitivity is (97.8%), and specificity is (98.8%). [ABSTRACT FROM AUTHOR]

The metaheuristics approach is important for solving many engineering applications, including healthcare. In the healthcare sector, many problems related to patients' treatments are not covered, and it is one of the issues for healthcare services across the globe. Healthcare support functions like clinical decision support, patient profiling, customized care, and health tracking could be the area where data analysis effectively provides end-user support. This paper focused on the metaheuristics algorithm concept with data mining technologies to provide a healthcare sector solution mechanism. This paper identifies the data analysis problem of the healthcare sector. It reviews various recent technologies used in healthcare and related sectors. It focused on the importance of metaheuristics algorithms, artificial intelligence, and implementation methodologies. [ABSTRACT FROM AUTHOR]

Copyright of Baghdad Science Journal is the property of Republic of Iraq Ministry of Higher Education & Scientific Research (MOHESR) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This paper examines a permutation flow-shop scheduling problem, which is a complex combinatorial problem encountered in many practical applications. The objective of the research is to reduce the maximum completion time, i.e., the makespan of all jobs. In order to increase productivity and to meet the demand, manufacturers are continuously under pressure to attain the shortest possible completion time. Estimation of accurate cycle time can tremendously help production planning and scheduling in manufacturing industries. Since production planning is characterised by NP-hardness and a wide range, traditional optimization methods and heuristic rules are unable to find satisfactory solutions. Q-learning, a type of reinforcement learning algorithm, is used in this paper to find a solution that is close to being optimal. Q-learning is a branch of machine learning referring to the way an intelligent agent should act in order to maximize the concept of cumulative reward in a given environment. To validate the performance of the algorithm, Taillard’s benchmark problems were solved and compared with the upper bound value. The results showed that the performance of the algorithm is better and has low computational time. Based on the performance of the proposed algorithm, two case studies were done and the solutions are compared with the performance of a metaheuristic algorithm. The result shows that the proposed algorithm can effectively and efficiently solve the problem stated above and that it is an interesting solution to resolving complex scheduling problems. [ABSTRACT FROM AUTHOR]

Training an Artificial Neural Network (ANN) algorithm is not trivial, which requires optimizing a set of weights and biases that increase dramatically with the increasing capacity of the neural network resulting in such hard optimization problems. Essentially, over recent decades, stochastic search algorithms have shown remarkable abilities for addressing hard optimization problems. On the other hand, pragmatically, abundant real-world problems suffer from the imbalance problem, where the distribution of data varies considerably among classes resulting in more training biases and variances which degrades the performance of the learning algorithm. This paper introduces three stochastic and metaheuristic algorithms for training the Multilayer Perceptron (MLP) neural network to solve the problem of imbalanced classifications. The utilized algorithms are the Grey Wolf Optimization (GWO), Particle Swarm Optimization (PSO), and the Salp Swarm Algorithm (SSA). The proposed GWO-MLP, PSO-MLP, and SSA-MLP are trained based on different objective functions; accuracy, f1-score, and g-mean. Whereas, it is evaluated based on 10 benchmark imbalanced datasets. The results show an advantage for f1-score, and g-mean fitness functions over the accuracy when the datasets are imbalanced. [ABSTRACT FROM AUTHOR]

One of the significant challenges in scaling agile software development is organizing software development teams to ensure effective communication among members while equipping them with the capabilities to deliver business value independently. A formal approach to address this challenge involves modeling it as an optimization problem: given a professional staff, how can they be organized to optimize the number of communication channels, considering both intra-team and inter-team channels? In this article, we propose applying a set of bio-inspired algorithms to solve this problem. We introduce an enhancement that incorporates ensemble learning into the resolution process to achieve nearly optimal results. Ensemble learning integrates multiple machine-learning strategies with diverse characteristics to boost optimizer performance. Furthermore, the studied metaheuristics offer an excellent opportunity to explore their linear convergence, contingent on the exploration and exploitation phases. The results produce more precise definitions for team sizes, aligning with industry standards. Our approach demonstrates superior performance compared to the traditional versions of these algorithms. [ABSTRACT FROM AUTHOR]

Undoubtedly, one of the most common machine learning challenges is multiclass classification. In light of this, a novel bio-inspired neural network (NN) has been developed to address multiclass classification-related issues. Given that weights and structure determination (WASD) NNs have been acknowledged to alleviate the disadvantages of conventional back-propagation NNs, such as slow training pace and trapping in a local minimum, we developed a bio-inspired WASD algorithm for multiclass classification problems (BWASDC) by using the metaheuristic beetle antennae search (BAS) algorithm to enhance the WASD algorithm's learning process. The BWASDC's effectiveness is then evaluated through applications in occupational classification systems. It is important to mention that systems of occupational classification serve as a fundamental indicator of occupational exposure. For this reason, they are highly significant in social science research. According to the findings of four occupational classification experiments, the BWASDC model outperformed some of the most modern classification models obtainable through MATLAB's classification learner app on all fronts. [ABSTRACT FROM AUTHOR]

In the optimization field, the ability to efficiently tackle complex and high-dimensional problems remains a persistent challenge. Metaheuristic algorithms, with a particular emphasis on their autonomous variants, are emerging as promising tools to overcome this challenge. The term "autonomous" refers to these variants' ability to dynamically adjust certain parameters based on their own outcomes, without external intervention. The objective is to leverage the advantages and characteristics of an unsupervised machine learning clustering technique to configure the population parameter with autonomous behavior, and emphasize how we incorporate the characteristics of search space clustering to enhance the intensification and diversification of the metaheuristic. This allows dynamic adjustments based on its own outcomes, whether by increasing or decreasing the population in response to the need for diversification or intensification of solutions. In this manner, it aims to imbue the metaheuristic with features for a broader search of solutions that can yield superior results. This study provides an in-depth examination of autonomous metaheuristic algorithms, including Autonomous Particle Swarm Optimization, Autonomous Cuckoo Search Algorithm, and Autonomous Bat Algorithm. We submit these algorithms to a thorough evaluation against their original counterparts using high-density functions from the well-known CEC LSGO benchmark suite. Quantitative results revealed performance enhancements in the autonomous versions, with Autonomous Particle Swarm Optimization consistently outperforming its peers in achieving optimal minimum values. Autonomous Cuckoo Search Algorithm and Autonomous Bat Algorithm also demonstrated noteworthy advancements over their traditional counterparts. A salient feature of these algorithms is the continuous nature of their population, which significantly bolsters their capability to navigate complex and high-dimensional search spaces. However, like all methodologies, there were challenges in ensuring consistent performance across all test scenarios. The intrinsic adaptability and autonomous decision making embedded within these algorithms herald a new era of optimization tools suited for complex real-world challenges. In sum, this research accentuates the potential of autonomous metaheuristics in the optimization arena, laying the groundwork for their expanded application across diverse challenges and domains. We recommend further explorations and adaptations of these autonomous algorithms to fully harness their potential. [ABSTRACT FROM AUTHOR]

This paper deals with strip packing metaheuristic algorithm selection using data mining techniques. Given a set of solved strip packing problem instances, the relationship between the instance characteristics and algorithm performance is learned and is used to predict the best algorithms to solve a new set of unseen problem instances. A framework capable of modelling this relationship for an automated packing algorithm selection is proposed. The effectiveness of the proposed framework is evaluated in the context of a large set of strip packing problem instances and the state-of-the-art strip packing algorithms. The results suggest a 91% accuracy in correctly identifying the best algorithm for a given instance. [ABSTRACT FROM AUTHOR]

Knowledge management (KM) is one of the main factors that have become extremely popular in recent years. KM is the processes which people explain information data using scientific and technological media and summarize it into concepts and rules to generate knowledge. This later can be implicit or explicit one. The aim of this contribution is to convert the tacit knowledge into explicit using Metaheuristics techniques. This paper aims to develop a model for converting tacit knowledge into explicit knowledge, using the Metaheuristics algorithm for the E-learning platform. For that purpose, the knowledge conversion process will respect the following steps: define the source of tacit knowledge and their methods, classify the tacit knowledge, then we evaluate the implicit knowledge conversion. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) system has confronted dramatic growth in high dimensionality and data traffic. The system named intrusion detection systems (IDS) is broadly utilized for the enhancement of security posture in an IT infrastructure. An IDS is a practical and suitable method for assuring network security and identifying attacks by protecting it from intrusive hackers. Nowadays, machine learning (ML)-related techniques were used for detecting intrusion in IoTs IDSs. But, the IoT IDS mechanism faces significant challenges because of physical and functional diversity. Such IoT features use every attribute and feature for IDS self-protection unrealistic and difficult. This study develops a Modified Metaheuristics with Weighted Majority Voting Ensemble Deep Learning (MM-WMVEDL) model for IDS. The proposed MM-WMVEDL technique aims to discriminate distinct kinds of attacks in the IoT environment. To attain this, the presented MM-WMVEDL technique implements min-max normalization to scale the input dataset. For feature selection purposes, the MM-WMVEDL technique exploits the Harris hawk optimization-based elite fractional derivative mutation (HHO-EFDM) technique. In the presented MM-WMVEDL technique, a Bi-directional long short-term memory (BiLSTM), extreme learning machine (ELM) and an ensemble of gated recurrent unit (GRU) models take place. A wide range of simulation analyses was performed on CICIDS-2017 dataset to exhibit the promising performance of the MM-WMVEDL technique. The comparison study pointed out the supremacy of the MM-WMVEDL method over other recent methods with accuracy of 99.67%. [ABSTRACT FROM AUTHOR]

Abstract: In their search for satisfactory solutions to complex combinatorial problems, metaheuristics methods are expected to intelligently explore the solution space. Various forms of memory have been used to achieve this goal and improve the performance of metaheuristics, which warranted the development of the Adaptive Memory Programming (AMP) framework [1]. This paper follows this framework by integrating Machine Learning (ML) concepts into metaheuristics as a way to guide metaheuristics while searching for solutions. The target metaheuristic method is Meta-heuristic for Randomized Priority Search (Meta-RaPS). Similar to most metaheuristics, Meta-RaPS consists of construction and improvement phases. Randomness coupled with a greedy heuristic are typically employed in the construction phase. While a local search heuristic is used in the improvement phase. This paper proposes adding a new learning phase, in which a ML method will be integrated. An Inductive Decision Tree (IDT) will be incorporated into the learning phase in an effort to learn from the information collected during the construction and improvement phases. The proposed approach will be demonstrated using instances for the Capacitated Vehicle Routing Problem (CVRP). [Copyright &y& Elsevier]

The Transmission Network Expansion Planning problem (TNEP) can be modeled either as a static, a pseudo-dynamic, or a dynamic problem. Most of the existing formulations do not include reactive power planning within the TNEP problem, leading to sub-optimal designs leading to higher system costs in reality. This paper proposes a dynamic (multi-stage) non-convex formulation that optimizes the addition of transmission circuits and reactive power compensation devices, accounting for operational costs including losses. The planning is done considering (N − 1) security constraints using an AC model. As there are no similar research works to benchmark the outcomes, the results were compared with those obtained from the static and pseudo-dynamic approaches, showing that the proposed approach provides more economical solutions. It is also shown that better solutions are obtained when Reactive Power Planning (RPP) is considered in the problem formulation. An improved Differential Evolution (DE) and Continuous Population Based Incremental Learning (PBILc) hybrid solution method (IDE-PBILc) is proposed which drastically improves calculation time and robustness. Comparisons with two different state-of-the-art metaheuristics are performed for validation. The results were obtained for the Garver 6-bus, IEEE 24-bus, and the IEEE 118-bus systems. Even though in this work uncertainties are not considered, the proposed approach could be of particular use when studying systems with high renewable energy penetration scenarios, due to its computational efficiency. • Dynamic AC transmission expansion planning. • Full AC non-convex modeling. • Reactive power planning. • (N − 1) Security constraints and operational costs. • Improved DE-PBILc solution method. [ABSTRACT FROM AUTHOR]

In today's era of data-driven digital society, there is a huge demand for optimized solutions that essentially reduce the cost of operation, thereby aiming to increase productivity. Processing a huge amount of data, like the Microarray based gene expression data, using machine learning and data mining algorithms has certain limitations in terms of memory and time requirements. This would be more concerning, when a dataset comes with redundant and non-important information. For example, many report-based medical datasets have several non-informative attributes which mislead the classification algorithms. To this end, researchers have been developing several feature selection algorithms that try to discard the redundant information from the raw datasets before feeding them to machine learning algorithms. Metaheuristic based optimization algorithms provide an excellent option to solve feature selection problems. In this paper, we propose a music-inspired harmony search (HS) algorithm based wrapper feature selection method. At the beginning, we use a chaotic mapping to initialize the population of the HS algorithm in order to better coverage of the search space. Further to complement the inferior exploitation of the HS algorithm, we integrate it with the Late Acceptance Hill Climbing (LAHC) method. Thus the combination of these two algorithms provides a good balance between the exploration and exploitation of the HS algorithm. We evaluate the proposed feature selection method on 15 UCI datasets and the obtained results are found to be better than many state-of-the-art methods both in terms of the classification accuracy and the number of features selected. To evaluate the effectiveness of our algorithm, we utilize a combination of precision, recall, F1 score, fitness value, and execution time as performance indicators. These metrics enable us to obtain a comprehensive assessment of the algorithm's abilities and limitations. We also apply our method on 3 microarray based gene expression datasets used for prediction of cancer to ensure the scalability and robustness as a feature selection method in real-life scenarios. In addition to this, we test our approach using the COVID-19 dataset, and it performs better than several metaheuristic based optimization techniques. • A novel metaheuristic based feature selection method is proposed. • Harmony Search Algorithm is combine with Late Acceptance Hill-Climbing method. • Chaotic map is used to for proper exploration of Harmony Search Algorithm. • Method is applied on many UCI and Microarray based Gene Expression datasets. • Achieved satisfactory results in terms of classification score and feature reduction. [ABSTRACT FROM AUTHOR]

Abstract Recurrent Neural Networks (RNNs) are possibly the most prevailing and advantageous type of neural network. On the other hand, these networks still have some weaknesses in terms of learning speed, error convergence, and accuracy due to long-term dependencies, which need to be solved. Long-term dependencies are mainly exploding and vanishing gradients through Back Propagation Learning Algorithm. In this paper, Long Short Term Memory or LSTM is used and well structured for resolving the above concerns. Four different optimizers based on Metaheuristic Algorithms are chosen to train LSTM (these are; Harmony Search (HS), Gray Wolf Optimizer (GWO), Sine Cosine (SCA), and Ant Lion Optimization algorithms (ALOA). The suggested representations are used for classification and analysis of real and medical time series data sets (Breast Cancer Wisconsin Data Set and Epileptic Seizure Recognition Data Set). Classification accuracy measure has been used instead of error rate and mean square error methods to train LSTM with above optimizing algorithms. The experimental results are verified using the 5-fold cross validation. Details of simulations and coding in R programing language can be obtained in the following link "https://github.com/pollaeng/rnn". [ABSTRACT FROM AUTHOR]

Post weld heat treatment, or PWHT, is often used to improve the mechanical properties of materials that have been welded. Several publications have investigated the effects of the PWHT process using experimental designs. However, the modeling and optimization using the integration of machine learning (ML) and metaheuristics have yet to be reported, which are fundamental steps toward intelligent manufacturing applications. This research proposes a novel approach using ML techniques and metaheuristics to optimize PWHT process parameters. The goal is to determine the optimal PWHT parameters for both single and multiple objective perspectives. In this research, support vector regression (SVR), K-nearest neighbors (KNN), decision tree (DT), and random forest (RF) were ML techniques employed to obtain a relationship model between PWHT parameters and mechanical properties: ultimate tensile strength (UTS) and elongation percentage (EL). The results show that the SVR demonstrated superior performance among ML techniques for both UTS and EL models. Then, SVR is used with metaheuristics such as differential evolution (DE), particle swarm optimization (PSO), and genetic algorithms (GA). SVR-PSO shows the fastest convergence among other combinations. The final solutions of single-objective and Pareto solutions were also suggested in this research. [ABSTRACT FROM AUTHOR]

The design of supply chain networks (SCNs) aims at determining the number, location, and capacity of production facilities, as well as the allocation of markets (customers) and suppliers to one or more of these facilities. This paper reviews the existing literature on the use of simulation-optimization methods in the design of resilient SCNs. From this review, we classify some of the many works in the topic according to factors such as their methodology, the approach they use to deal with uncertainty and risk, etc. The paper also identifies several research opportunities, such as the inclusion of multiple criteria (e.g., monetary, environmental, and social dimensions) during the design-optimization process and the convenience of considering hybrid approaches combining metaheuristic algorithms, simulation, and machine learning methods to account for uncertainty and dynamic conditions, respectively. [ABSTRACT FROM AUTHOR]

In data mining and machine learning, feature selection is a critical part of the process of selecting the optimal subset of features based on the target data. There are 2n potential feature subsets for every n features in a dataset, making it difficult to pick the best set of features using standard approaches. Consequently, in this research, a new metaheuristics-based feature selection technique based on an adaptive squirrel search optimization algorithm (ASSOA) has been proposed. When using metaheuristics to pick features, it is common for the selection of features to vary across runs, which can lead to instability. Because of this, we used the adaptive squirrel search to balance exploration and exploitation duties more evenly in the optimization process. For the selection of the best subset of features, we recommend using the binary ASSOA search strategy we developed before. According to the suggested approach, the number of features picked is reduced while maximizing classification accuracy. A ten-feature dataset from the University of California, Irvine (UCI) repository was used to test the proposed method's performance vs. eleven other state-of-the-art approaches, including binary grey wolf optimization (bGWO), binary hybrid grey wolf and particle swarm optimization (bGWO-PSO), bPSO, binary stochastic fractal search (bSFS), binary whale optimization algorithm (bWOA), binary modified grey wolf optimization (bMGWO), binary multiverse optimization (bMVO), binary bowerbird optimization (bSBO), binary hybrid GWO and genetic algorithm (bGWO-GA), binary firefly algorithm (bFA), and bGAmethods. Experimental results confirm the superiority and effectiveness of the proposed algorithm for solving the problem of feature selection. [ABSTRACT FROM AUTHOR]

In recent years, continuous metaheuristics have been a trend in solving binary-based combinatorial problems due to their good results. However, to use this type of metaheuristics, it is necessary to adapt them to work in binary environments, and in general, this adaptation is not trivial. The method proposed in this work evaluates the use of reinforcement learning techniques in the binarization process. Specifically, the backward Q-learning technique is explored to choose binarization schemes intelligently. This allows any continuous metaheuristic to be adapted to binary environments. The illustrated results are competitive, thus providing a novel option to address different complex problems in the industry. [ABSTRACT FROM AUTHOR]

The proper configuration of a metaheuristic requires specific and advanced knowledge, for instance to decide an appropriate population size, to effectively set probability parameters, or to define a correct termination criterion. During the last years there is a trend to let the metaheuristic to autonomously self-control the internal configuration. In this way, by using learning mechanisms, the algorithm is able to learn from the solving process and automatically adapt its configuration in order to improve their performance. Following this research trend, in this paper we explore the ability to autonomously predict population size and termination criteria. The population prediction allows the metaheuristic to dynamically vary the population size during solving time. In this regard, a suitable number of agents is used for each part of the process in order to efficiently explore the space of solutions. On the other hand, predicting the termination criteria is useful as well, the smart usage of resources in the optimization field can bring several advantages in solving complex industrial problems. In this regard, since the number of iterations is estimated during solving time, the proposed architecture avoids useless processing time that does not lead to any solution improvement. We illustrate interesting results using the spotted hyena optimizer and crow search algorithm solving 24 benchmarks functions and the multidimensional knapsack problem, where the proposed approach notably competes with several state-of-the-art optimization algorithms. • Robust hybrid architecture to tackle optimization problems. • A key issue in population-based approaches: Adapting population size on run-time. • A key issue in metaheuristic approaches: Adapting the termination criteria on run-time. • Scalability (module 1), movement operators from different algorithms can be employed. • Scalability (modules 3, 4), incorporation of multiple machine learning methods. [ABSTRACT FROM AUTHOR]

This paper contributes to the intersection of operations research and machine learning in the context of autonomous ridesharing. In this work, autonomous ridesharing operations are reproduced through an event-based simulation approach and are modeled as a sequence of static subproblems to be optimized. The optimization framework consists of a novel data-driven metaheuristic within a two phase approach. The first phase consists of a greedy insertion heuristic that assigns new online requests to vehicles. The second phase consists of a local-search based metaheuristic that iteratively revisits previously-made vehicle-trip assignments through intra- and inter-vehicle route exchanges. These exchanges are performed by selecting from a pool of destroy–repair operators using a machine learning approach that is trained offline on a large dataset composed of more than one and a half million examples of previously-solved autonomous ridesharing subproblems. Computational results are performed on multiple dynamic instances extracted from real ridesharing data published by Uber Technologies Inc. Results show that the proposed machine learning-based optimization approach outperforms benchmark state-of-the-art data-driven metaheuristics by up to about nine percent, on average. Managerial insights highlight the correlation between selected vehicle routing features and the performance of the metaheuristics in the context of autonomous ridesharing operations. • Tackling an autonomous ridesharing problem with dynamic demand. • Developing an event-based simulation approach. • Developing a machine learning-based large neighborhood search. • Training machine learning models on hundreds of thousands of neighborhood move examples. • Determining most influential vehicle routing features which impact the search. [ABSTRACT FROM AUTHOR]

Artificial Intelligence (AI) encompasses various domains such as Machine Learning (ML), Deep Learning (DL), and other cognitive technologies which have been widely applied in healthcare sector. AI models are utilized in healthcare sector in which the machines are used to investigate and make decisions based on prediction and classification of input data. With this motivation, the current study involves the design of Metaheuristic Optimization with Kernel Extreme Learning Machine for COVID19 Prediction Model on Epidemiology Dataset, named MOKELM-CPED technique. The primary aim of the presented MOKELM-CPED model is to accomplish effectual COVID-19 classification outcomes using epidemiology dataset. In the proposed MOKELM-CPED model, the data first undergoes pre-processing to transform the medical data into useful format. Followed by, data classification process is performed by following Kernel Extreme Learning Machine (KELM) model. Finally, Symbiotic Organism Search (SOS) optimization algorithm is utilized to fine tune the KELM parameters which consequently helps in achieving high detection efficiency. In order to investigate the improved classifier outcomes of MOKELM-CPED model in an effectual manner, a comprehensive experimental analysis was conducted and the results were inspected under diverse aspects. The outcome of the experiments infer the enhanced performance of the proposed method over recent approaches under distinct measures. [ABSTRACT FROM AUTHOR]

Scalability of optimization algorithms is a major challenge in coping with the ever-growing size of optimization problems in a wide range of application areas from high-dimensional machine learning to complex large-scale engineering problems. The field of large-scale global optimization is concerned with improving the scalability of global optimization algorithms, particularly, population-based metaheuristics. Such metaheuristics have been successfully applied to continuous, discrete, or combinatorial problems ranging from several thousand dimensions to billions of decision variables. In this two-part survey, we review recent studies in the field of large-scale black-box global optimization to help researchers and practitioners gain a bird’s-eye view of the field, learn about its major trends, and the state-of-the-art algorithms. Part I of the series covers two major algorithmic approaches to large-scale global optimization: 1) problem decomposition and 2) memetic algorithms. Part II of the series covers a range of other algorithmic approaches to large-scale global optimization, describes a wide range of problem areas, and finally, touches upon the pitfalls and challenges of current research and identifies several potential areas for future research. [ABSTRACT FROM AUTHOR]

Population-based approaches have given us new search strategies and ideas in order to solve optimization problems. Usually, these methods are based on the performance carried out by a finite number of agents, which by the interaction between them they evolve and work all over the search space. Also, it is well-known that the correct employment of parameter values in this kind of method can positively impact their performance and behavior. In this context, the present work focuses on the design of a hybrid architecture which smartly balances the population size on run-time. In order to smartly balance and control the population size, a modular approach, named Linear Modular Population Balancer (LMPB), is proposed. The main ideas behind the designed architecture include the solving strategy behind a population-based metaheuristic, the influence of learning components based on multiple statistical modeling methods which transform the dynamic data generated into knowledge, and the possibilities to tackle both discrete and continuous optimization problems. In this regard, three modules are proposed for LMPB, which concern tasks such as the management of the population-based algorithm, parameter setting, probabilities, learning methods, and selection mechanism for the population size to employ. In order to test the viability and effectiveness of our proposed approach, we solve a set of well-known benchmark functions and the multidimensional knapsack problem (MKP). Additionally, we illustrate promising solving results, compare them against state-of-the-art methods which have proved to be good options for solving optimization problems, and give solid arguments for future work in the necessity to keep evolving this type of proposed architecture. [ABSTRACT FROM AUTHOR]

During the past fewyears, research in applying machine learning (ML) to design efficient, effective, and robust metaheuristics has become increasingly popular. Many of those machine learning-supported metaheuristics have generated high-quality results and represent state-of-the-art optimization algorithms. Although various appproaches have been proposed, there is a lack of a comprehensive survey and taxonomy on this research topic. In this article, we will investigate different opportunities for using ML into metaheuristics. We define uniformly the various ways synergies that might be achieved. A detailed taxonomy is proposed according to the concerned search component: target optimization problem and low-level and high-level components of metaheuristics. Our goal is also to motivate researchers in optimization to include ideas from ML into metaheuristics. We identify some open research issues in this topic that need further in-depth investigations. [ABSTRACT FROM AUTHOR]

The Artificial Neural Network (ANN) is an important machine learning tool used in medical data classification for disease diagnosis. The learning algorithm in ANN training plays a crucial role in classification performance. Various approaches have been successfully applied as a learning algorithm for ANN training. This paper performs an experimental study that investigates the performance of different metaheuristics as learning algorithms to train the ANN for medical data classification tasks. The experiments are carried out on 15 well-known medical datasets. A comparative study is conducted with the classical Levenberg–Marquardt (LM) and other thirteen recent and relevant metaheuristics. Different evaluation criteria such as accuracy, sensitivity, specificity, precision, Geometric Mean, F-Measure, false-positive rate (FPR) are considered for performance estimation. The classification results are analyzed using Multi-Criteria Decision Making (MCDM) method, and the results with analysis establish that the Equilibrium Optimizer algorithm outperforms all the other algorithms included in the comparative study. • Artificial Neural Network training using metaheuristic algorithms. • An experimental study in medical data classification. • Performance analysis using multi-criteria decision making. • Equilibrium Optimizer shows superior performance over the competitive algorithms. [ABSTRACT FROM AUTHOR]

We introduce new diversification methods for zero-one optimization that significantly extend strategies previously introduced in the setting of metaheuristic search. Our methods incorporate easily implemented strategies for partitioning assignments of values to variables, accompanied by processes called augmentation and shifting which create greater flexibility and generality. We then show how the resulting collection of diversified solutions can be further diversified by means of permutation mappings, which equally can be used to generate diversified collections of permutations for applications such as scheduling and routing. These methods can be applied to non-binary vectors using binarization procedures and by diversification-based learning procedures that provide connections to applications in clustering and machine learning. Detailed pseudocode and numerical illustrations are provided to show the operation of our methods and the collections of solutions they create. [ABSTRACT FROM AUTHOR]

Currently, industry is undergoing an exponential increase in binary-based combinatorial problems. In this regard, metaheuristics have been a common trend in the field in order to design approaches to successfully solve them. Thus, a well-known strategy includes the employment of continuous swarm-based algorithms transformed to perform in binary environments. In this work, we propose a hybrid approach that contains discrete smartly adapted population-based strategies to efficiently tackle binary-based problems. The proposed approach employs a reinforcement learning technique, known as SARSA (State–Action–Reward–State–Action), in order to utilize knowledge based on the run time. In order to test the viability and competitiveness of our proposal, we compare discrete state-of-the-art algorithms smartly assisted by SARSA. Finally, we illustrate interesting results where the proposed hybrid outperforms other approaches, thus, providing a novel option to tackle these types of problems in industry. [ABSTRACT FROM AUTHOR]