Search

Introducing a novel meta-heuristic optimization algorithm, the Flood Algorithm (FLA) draws inspiration from the intricate movement and flow patterns of water masses during flooding events in river basins. FLA mathematically models key phenomena such as the movement of water toward slopes, flow rates over time, soil permeability effects, and periodic increases and decreases in water levels from precipitation and losses. Leveraging these models, the algorithm guides the movement and evolution of a population of potential solutions toward enhanced optimality. The algorithm endeavors to establish an appropriate correlation between the fundamental aspects of natural flood events and the optimization process. Its formulation and working mechanism are described in detail. It operates in two main phases—a regular movement phase, where the population moves naturally toward current best solutions, and a flooding phase, which introduces random disturbances to increase diversity. New solutions are periodically introduced while weaker ones are removed, mirroring the natural cycles of water levels. FLA's effectiveness is demonstrated through its application on well-known benchmark optimization problems and engineering design problems. Extensive comparisons have been carried out on CEC2005 functions using 16 algorithms in both basic and enhanced modes, as well as on CEC2014 functions with dimensions 30, 50, and 100 using a total of 20 other algorithms. These rigorous studies unequivocally confirm the robustness and strength of the proposed algorithm. Furthermore, the algorithm's performance on 12 constrained engineering problems demonstrates its ability to tackle real-world challenges. The FLA's source code is publicly available at https://www.optim-app.com/projects/fla. [ABSTRACT FROM AUTHOR]

The efficacy of meta-heuristic algorithms has been demonstrated in solving unconstrained optimization problems. Inspired by the food foraging behavior of beetles, we propose herein a model-free meta-heuristic optimization algorithm, referred to as the Beetle Antennae Search-Bellwether Swarm, which incorporates the schemes of adaptive antenna fiber length and direction vector. A joint annealing-heating scheme along with the re-initialization mechanism is proposed such that each antenna fiber length could simultaneously gradually decrease or increase in different circumstances. We also propose to adapt each individual antenna fiber length based on the best smell perception direction (bellwether). The proposed schemes of adaptive antenna fiber length yield the algorithm almost insusceptible to different initial antenna fiber lengths for a relatively large range of initial antenna fiber lengths. The softmax combiner is leveraged to obtain the direction vector such that the beetle would be steered to step toward the new randomly combined direction and arrive at a candidate position. The beetle position is further updated by evaluating whether the candidate position is conducive to minimizing the objective function. The illustrative simulations demonstrate the fast initial convergence, cost-effectiveness and feasibility of the proposed algorithm. [ABSTRACT FROM AUTHOR]

High utility itemset mining is an essential new task in data mining, which is obtained from the extension of frequent itemset mining problems. The difference between these two groups is that in high utility itemset mining, the utilities of the items are essential in addition to the number of items. To find the itemsets with more utilities, defining a minimum utility threshold is necessary by testing different solutions to obtain the desired outcome. This is challenging and needs more time to reach the appropriate value. In this article, we propose a method called TKU-BChOA to explore top-k high utility itemsets according to the number of required itemsets based on the Chimp meta-heuristic algorithm. Because, in comparison with the meta-heuristic methods, the deterministic methods require more time and memory. Experimental results on five datasets show that the proposed approach outperforms the previous deterministic methods in memory usage. The proposed method performs better in terms of time and memory on large datasets than other approximate methods and is more accurate. [ABSTRACT FROM AUTHOR]

Snake optimizer (SO) is an optimization algorithm drawn from the reproductive habits of serpents. It exhibits outstanding effectiveness in solving continuous optimization problems. However, SO may face some performance challenges related to its population diversity and early convergence behavior. In this paper, we address the challenges of SO by introducing the Hybrid snake optimizer algorithm (HSOA). HSOA is a novel approach to optimization that incorporates two new optimization techniques into the SO algorithm. First, it incorporates a new opposition-based learning technique called Oppositional-mutual learning into the initialization stage of the SO algorithm. Second, it integrates dynamic polynomial mutation, which is an intelligent mutation method, into the initialization and optimization stages of the SO algorithm. These integrated approaches aim to increase the population's diversity of SO, while improving its searchability during its optimization stage. In power systems, the economic load dispatch (ELD) is an intricate optimization problem that becomes more challenging when the restrictions of the valve point effect (VPE) are incorporated into it. ELD with VPE is non-convex, lacking smoothness, and exhibiting nonlinearity that considers operational limitations expressed as both equality and inequality constraints to generate electricity. The suggested HSOA algorithm underwent evaluation and was compared with 47 renowned optimization algorithms across five real-world ELD problems with different specifications: generators with different unit capacities, transmission losses, prohibited operation zones, and Ramp Rate restrictions. The experimental results demonstrate that HSOA produces competitive solutions for the five real-world ELD problems. In detail, HSOA achieves the top rank in three cases of ELD problems with a 3-unit generator, and it secures the second and third positions in high-dimensional ELD problems with 40-unit and 80-unit generators, respectively. The statistical tests confirm the reliability and efficiency of HSOA. In addition, the effectiveness of HSOA was evaluated using the single-objective IEEE-CEC 2014 functions and compared to the results of eight popular metaheuristic algorithms. The results demonstrate that HSOA is a competitive optimization algorithm capable of solving the functions of IEEE-CEC 2014. [ABSTRACT FROM AUTHOR]

This paper proposes a new meta-heuristic optimization algorithm called Cleaner Fish Optimization algorithm (CFO) inspired by cleaner fish. The CFO simulates the movement behavior of cleaner fish when performing "cleaning services" and the behavior of female may change its sex to become a male, and defines two modes of position update. In addition, a two-generation cycle operation strategy is proposed to realize the optimization process. To verify the effectiveness of the CFO algorithm, 23 well-known CEC benchmark functions, CEC-2017 benchmark functions and 4 engineering design problems are adopted. Simulation results show that the proposed algorithm has a faster convergence rate and better optimal solution when it compared with several other meta-heuristic algorithms. [ABSTRACT FROM AUTHOR]

Child Drawing Development Optimization is a recently developed metaheuristic algorithm that has been demonstrated to perform well on multiple benchmark tests. In this paper, a binary Child Drawing Development Optimization (BCDDO) is proposed for wrapper feature selection. The proposed BCDDO is utilized to choose a subset of important features to reach the highest classification accuracy. Harris Hawk optimization, salp swarm algorithm, gray wolf optimization, and whale optimization algorithm are utilized to evaluate the effectiveness and efficiency of the suggested feature selection method. In the field of feature selection to improve classification accuracy, the proposed method has gained a considerable classification accuracy advantage over previously mentioned methods. Four datasets are used in this research work; breast cancer, moderate COVID, big COVID, and Iris using XGBoost classifier and the classification accuracies were (98.83%, 98.75%, 99.36%, and 96%), respectively, for the four mentioned datasets. [ABSTRACT FROM AUTHOR]

The intrusion detection system is responsible for revealing different intrusion activities, including the denial of service, man-in-middle, Mirai, Scan, and other types of intrusion activities. It is used in many applications, including the smart home Internet of Things networks, where security risks threaten the privacy of individuals. In this context, many works were proposed for detecting and classifying the different types of attacks. However, many challenges are identified for this type of problem, such as the large amount of data available, the imbalanced nature of the data, and the quality of detection and classification outcomes. This paper aims to address these challenges by proposing an approach that considers a metaheuristic-based random weight neural network to detect intrusion activities and classify the different types and subtypes of activities. The following points summarize the contribution of this paper. First, the automatic tuning of the neural network parameters where the weights, biases, regularization value, the number of neurons, and the type of activation function are optimized by different metaheuristic algorithms to produce high-quality results. Second, the proposed approach adopts a clustering with reduction technique to tackle the challenge of processing large volumes of data. Third, oversampling the dataset is also embedded in the proposed approach to avoid a biased classification of the majority class. The experiments are conducted based on a large dataset with more than half a million instances. The results show that the proposed approach outperforms the other classification approaches in geometric mean (G-Mean) and has promising results. [ABSTRACT FROM AUTHOR]

In artificial intelligence (AI), optimization techniques are used to solve several problems in different fields. One of these areas is software reliability verification, which is an important part of software products, as it helps determine how reliable the software is to complete its functions. This is done by estimating the parameters of software reliability growth models (SRGMs). SRGMs predict the expected number of failures after completion, while also serving as an indicator of software readiness for delivery. Therefore, this study aims to optimize the estimation of these parameters based on the available failure data using one of the stochastic optimization algorithms, the simulated annealing algorithm (SA) due to its power and effectiveness. Three SRGMs' models are studied: delayed S-shaped, Musa-Okumoto logarithmic and Power models, to examine the feasibility of the proposed algorithm using five different data sets. The results were compared and analyzed with several algorithms: Particle swarm optimization (PSO), cuckoo search (CS), modify whale optimization algorithm (MWOA), S-shaped model with logistic TEF and social spider algorithm (SSA). A comparison was also made with recent SRGMs that do not rely on AI techniques. The results showed that the proposed algorithm based on SA outperformed all other methods in finding the optimal parameters. [ABSTRACT FROM AUTHOR]

This paper proposes a bit-level cryptographic model for encrypting medical signals, patient information, clinical data, and images of various modalities. The model has utilized chaotic maps due to properties like non-periodicity, nonlinearity, ergodicity, deterministic, unpredictable, and sensitive to the initial conditions. Three chaotic maps are employed in a chained sequence at different levels of the model to elevate the randomness and provide uniform key distribution with larger key space. Also, the self-adaptive bald eagle search optimization algorithm has been employed to enhance the properties of chaotic maps by optimizing the initialization and control parameters. The cyclic redundancy check approach is involved to detect any change in the data during the transmission at the receiver side along with circular shift to make encrypted data resilient of differential attack. Therefore, the encryption model boosts the unpredictability, security, and protection against a wide range of threats. The encryption model performance efficiency is demonstrated using different types of security analysis to show resistance against statistical, entropy, differential, cropping, and brute-force attacks. [ABSTRACT FROM AUTHOR]

The gazelle optimization algorithm (GOA) is an iterative optimization method inspired by the agile movements of gazelles, employing adaptive step sizes and velocity adjustments for rapid convergence in continuous search spaces. However, GOA tends to lack diversity, leading to issues like local minima trapping and premature convergence. This paper addresses these limitations by introducing dynamic opposition-based learning (OBL) and incorporating a balanced sine-control-logistic chaotic mapping system, resulting in the improved GOA (IGOA). Dynamic OBL accelerates the search process, improving learning and selecting superior candidate solutions, while chaotic mapping in chaotic local search widens ranges for exploration and local optima escape. Evaluating IGOA against seven algorithms, including GOA and others, across 31 general test functions, the results consistently showcase IGOA's superior efficiency in achieving solutions closest to optima, early convergence, and hit rate when compared to alternative algorithms. [ABSTRACT FROM AUTHOR]

Topology control is a significant research problem when it comes to designing energy-efficient wireless sensor networks (WSNs). It is a metric for evaluating network service quality. To ensure the quality of network services, it is critical to provide energy-efficient sensor node scheduling in order to extend the network's lifespan. In this paper, we propose an energy-aware scheduling protocol (ESP)-based hybrid metaheuristic technique for minimizing the amount of energy consumption and ensuring sufficient coverage for the monitored area while maximizing the network lifespan for WSNs. The ESP consists of two phases: distributed clustering and sensor node scheduling. The clustering phase implements the DBSCAN clustering algorithm with slide modification to cluster the sensor nodes in the area of interest into clusters of nodes, and the efficient periodic cluster head election approach is proposed based on criteria such as remaining energy, number of neighbors, and distance for each node in the cluster. This clustering is only performed at the beginning to group the nodes into clusters. Then, only the cluster head election will be done every period. In the scheduling phase, we model the scheduling optimization problem using three objectives: reducing the number of uncovered zones, minimizing the number of active sensors, and selecting the active sensor nodes with the maximum remaining energy. This scheduling optimization model is solved using a hybrid metaheuristic algorithm. When the genetic algorithm's operators were added to the regular Cuckoo algorithm, the balance between exploration and exploitation abilities was improved even more. The algorithms were able to search a larger area. Extensive simulation experiments were conducted using the OMNeT++ network simulator. The results show that the proposed ESP protocol introduces better performance in terms of coverage ratio, active sensor ratio, energy consumption, network lifespan, and execution time compared to other existing techniques. [ABSTRACT FROM AUTHOR]

This paper proposes the Love Evolution Algorithm (LEA), a novel evolutionary algorithm inspired by the stimulus–value–role theory. The optimization process of the LEA includes three phases: stimulus, value, and role. Both partners evolve through these phases and benefit from them regardless of the outcome of the relationship. This inspiration is abstracted into mathematical models for global optimization. The efficiency of the LEA is validated through numerical experiments with CEC2017 benchmark functions, outperforming seven metaheuristic algorithms as evidenced by the Wilcoxon signed-rank test and the Friedman test. Further tests using the CEC2022 benchmark functions confirm the competitiveness of the LEA compared to seven state-of-the-art metaheuristics. Lastly, the study extends to real-world problems, demonstrating the performance of the LEA across eight diverse engineering problems. Source codes of the LEA are publicly available at https://ww2.mathworks.cn/matlabcentral/fileexchange/159101-love-evolution-algorithm. [ABSTRACT FROM AUTHOR]

This paper proposes a novel hyper-heuristic algorithm termed evolutionary multi-mode slime mold optimization (EMSMO) for addressing continuous optimization problems. The architecture of a typical hyper-heuristic algorithm comprises two main components: the high-level component and the low-level component. The low-level component contains a set of low-level heuristics (LLHs) and intrinsic problem attributes, while the high-level component manipulates the LLHs to construct the sequence of heuristics. Inspired by the foraging behaviors of slime mold, we designed four easy-implemented search strategies including the search for food, approach food, wrap food, and re-initialization as the LLHs for the low-level component. In the high-level component, we adopt an improvement-based probabilistic selection function that contains two metrics: (1) the probability of improvement and (2) the normalized improvement. The selection function cooperates with the roulette wheel strategy to construct the optimization sequence. To evaluate the performance of our proposal, we implement comprehensive numerical experiments on CEC2013 benchmark functions and three engineering optimization problems. Six classic or advanced evolutionary algorithms and three hyper-heuristic algorithms are applied as competitor algorithms to evaluate the competitiveness of EMSMO. Experimental and statistical results show that EMSMO has broad prospects for solving continuous optimization problems. [ABSTRACT FROM AUTHOR]

The software-defined network paradigm, ensembled with a wireless sensor network, has emerged as a promising phenomenon to enable multi-tasking, re-configuration, and scalability. Termed the software-defined wireless sensor network (SDWSN), it divides the network into two planes: data and control. The data plane consists of software-defined sensor nodes (SDSN) that sense monitoring activity and generate data. On the other hand, the control plane has controller/control Nodes (CN) which collect data from SDSN, perform data aggregation, and then transmit toward the sink node. These CNs consume more amount of energy as compared to SDSNs as they perform multiple tasks. Following this scenario, this paper proposes an energy-efficient multi-objective optimization approach to solve the CNs placement problem through a meta-heuristic algorithm by considering the nodes' location, energy, and load distribution. This paper presents a particle swarm optimization-based controller placement and re-placement (PSO-CPR) algorithm for SDWSN. The PSO-CPR elects SDSNs to become CN based on their distance, residual energy, and capacity in the network. Moreover, the placement of a CN rotates within the cluster to avoid its failure and balance energy consumption. The simulation results show improved CN placement with respect to the state-of-the-art algorithms in terms of average delay by 23.5–37.4%, energy consumption by 18.6–32.6%, and probabilistic load distribution by 17.7–54.1%. Moreover, the comparative study also indicates that PSO-CPR achieve promising result by reducing packet loss by 14.4–27.5% and network re-clustering period by 32.3–68.3% and enhance the network lifetime by 22.6–42.5%. [ABSTRACT FROM AUTHOR]

Taking a lot of time to solve optimization problems has become a challenge for metaheuristic algorithms. Due to independence of the metaheuristics components, parallel processing is a good option to reduce the computational time and to find high quality solutions that are close to the optimum with an acceptable cost. One of these metaheuristic algorithms is the Sailfish Optimizer (SFO) which is inspired by a group of hunting sailfish. The SFO algorithm uses a simple method to provide a dynamic balance between exploration and exploitation phases, creates a swarm diversity, avoids local optima, and guarantees high convergence speed. It has been shown that the SFO algorithm outperforms various state-of-art metaheuristic algorithms for multimodal and high dimensional benchmark functions and complicated real-world optimization problems in terms of accuracy and speed by CPU and GPU implementation. In this paper, to speedup this algorithm and increase its performance we propose a reconfigurable hardware version of SFO implemented on Field Programmable Gate Array (FPGA). The FPGA-based SFO can be a very good option in many applications with massive calculations. Due to the inherent parallelism and high computing capabilities of FPGA, the SFO algorithm gains optimum computational time despite the complexity of optimization problems. We have compared the performance of the proposed FPGA-based SFO with its CPU and GPU implementation and some other metaheuristic algorithms. The results show the FPGA implementation is considerably faster than the CPU and GPU implementation. Also, it outperforms other compared FPGA-based metaheuristic algorithms in terms of execution time and convergence speed. [ABSTRACT FROM AUTHOR]

Clustering is a popular data analysis technique that can explore the structure of data through cluster analysis. Similar data are put into the same cluster, while dissimilar data allocate to other clusters. The similarity/dissimilarity among data objects is determined using a distance function. Further, clustering algorithms aim to choose the optimal set of centroids for obtaining better partitioning, but clustering accuracy is always susceptible. This issue of clustering is addressed through meta-heuristic algorithms. This research also aims to handle the accuracy issue and presents a new algorithm for effective cluster analysis. The proposed clustering algorithm is inspired by a water flow optimizer (WFO). The WFO algorithm performance is validated on the well-defined clustering problems based on SSE, accuracy (AR) and detection rate (DR) parameters. The results indicate that the WFO algorithm gets higher clustering results in terms of SSE, AR and DR than the same class of algorithms. The performance is also validated using Friedman statistical test followed by a post hoc test. Results indicated that the proposed WFO gets better statistical results than other clustering algorithms. [ABSTRACT FROM AUTHOR]

Single-solution-based optimization algorithms are computationally cheap yet powerful methods that can be used on various optimization tasks at minimal processing expenses. However, there is a considerable shortage of research in this domain, resulting in only a handful of proposed algorithms over the last four decades. This study proposes the Prism Refraction Search (PRS), a novel, simple yet efficient, single-solution-based metaheuristic algorithm for single-objective real-parameter optimization. PRS is a physics-inspired algorithm modeled on a well-known optimization paradigm in ray optics arising from the refraction of light through a triangular prism. The key novelty lies in its scientifically sound background that is supported by the well-established laws of physical optics. The proposed algorithm is evaluated on several numerical objectives, including 23 classical benchmark functions, the CEC-2017 test suite, and five standard real-world engineering design problems. Further, the results are analyzed using standard statistical tests to prove their significance. Extensive experiments and comparisons with state-of-the-art metaheuristic algorithms in the literature justify the robustness and competitive performance of the PRS algorithm as a lightweight and efficient optimization strategy. [ABSTRACT FROM AUTHOR]

The advent of Internet of Things (IoT) in agriculture has revolutionized the way farmers monitor and manage their crops. IoT-enabled sensors can provide real-time data on various environmental parameters such as temperature, humidity, soil moisture, and crop growth, which can be used to make informed decisions and optimize crop yield. However, the vast amount of data generated by these sensors poses a significant challenge in terms of data processing and communication. To address this challenge, clustering is often used to group the sensors into clusters and elect a cluster head (CH) to communicate with the gateway node. The selection of an appropriate CH and the optimal path for data transmission are critical factors that affect the performance of the IoT system. In this paper, we propose a novel approach to optimize the CH selection and path selection using modified fuzzy logic, Whale optimization algorithm (WOA) and enhanced crow swarm optimization (ECSO). Fuzzy logic is used to evaluate the relevant parameters such as energy, distance, overhead, trust, and node degree to select the most suitable CH. ECSO is then employed to find the optimal path for data transmission based on the selected CH. We evaluate the proposed approach using simulation experiments in a smart agriculture scenario. The results show that the proposed approach outperforms existing approaches in terms of throughput, packet delivery ratio, delay, and energy efficiency. The proposed system demonstrates substantial advantages in terms of throughput, with a remarkable throughput value of 68 Mbps, outperforming the other methods, including EQSR (12.5 Mbps), E2S-DRL (60 Mbps), and OSEAP. The packet delivery ratio (PDR) is also significantly higher for the proposed system, reaching 90.9%. The proposed approach can significantly improve the performance of IoT-enabled smart agriculture systems, leading to better crop yield and higher profitability for farmers. The results of the simulation experiments demonstrate the superiority of the proposed approach over the existing one's throughput, packet delivery ratio (PDR), delay, and energy consumption efficiency is found in the result section. [ABSTRACT FROM AUTHOR]

Several real-world scientific and industrial workflow applications adopt elastic and cost-efficient cloud services to fulfill their requirement. There are two stakeholders in the system, namely the user and the provider each of which tries to maximize its profit and at the same time minimize their possible overall costs. Since task scheduling algorithms for workflow executions determine which task should be possibly executed on what resources, they have a drastic impact on both the user's quality of experiences and on the underlying resource utilization. Indeed, an efficient task scheduling algorithm can meet service-level agreement (SLA) for both sides. For the sake of the importance of the issue, this survey presents a subjective taxonomy on the task scheduling schemes in the literature for workflow executions in cloud computing environments to be a guideline for future improvement. It classifies the literature based on the proposed algorithms in the literature, objectives, stakeholders' requirements, and evaluation metrics. This survey highlights research trends, challenges, research gaps, potential solutions, and future direction. It can also pave the way for further processing, improvement and strengthening of existing approaches or devising novel ones for interested researchers in the field of task scheduling problems. [ABSTRACT FROM AUTHOR]

Arithmetic optimization algorithm (AOA) is a meta-heuristic optimization method based on mathematical operators proposed in recent years. Although it has good performance, it can also lead to insufficient local search ability and falling into local optima when solving complex optimization problems. In order to make up for the above shortcomings, the optimization performance of AOA is further improved. This paper proposes a hybrid algorithm based on AOA and particle swarm optimization (PSO) called HAOAPSO. Firstly, a compound opposition-based learning (COBL) strategy is introduced to broaden the scope of finding optimal solutions to help the algorithm better jump out of local optima. Secondly, PSO is combined with AOA that integrates COBL to improve the algorithm's local search ability, so as to improve the overall search efficiency of the algorithm. In addition, experiments are performed on 23 classical benchmark functions with different characteristics and five engineering design optimization problems, and the experimental results of HAOAPSO are compared with those of other well-known optimization algorithms to comprehensively evaluate the performance of the proposed algorithm. The simulation results show that HAOAPSO can provide better solutions in most cases when solving global optimization problems such as engineering, with better convergence speed and accuracy. [ABSTRACT FROM AUTHOR]

The utilization of cloud computing environments is highly popular for carrying out workflow executions due to its ability to provide clients with immediate access to computing resources. Among the various workflow scheduling problems in the cloud, deadline-constrained workflow scheduling has garnered increasing attention in recent years. This paper introduces a hybrid scheduling algorithm known as Partial Critical Path–Ant Colony Optimization (PCP–ACO), which aims to minimize the execution cost of a workflow while ensuring that it meets the user-defined deadline in cloud environments. PCP–ACO is a list scheduling algorithm that combines the PCP heuristic algorithm with the meta-heuristic ACO to achieve faster convergence. The list scheduling algorithm consists of two phases: task ordering and resource selection. In the case of PCP–ACO, the first step involves calculating a topological sort of the workflow tasks to assign priority to each task. Subsequently, the ACO meta-heuristic is employed to allocate the appropriate resource to each task of the workflow, based on their respective sub-deadlines that are computed using the PCP heuristic. In order to evaluate the effectiveness of the proposed algorithm, several experiments were conducted using five real scientific workflows. The results demonstrate that PCP–ACO outperforms the IC-PCP, L-ACO, and HP-GA algorithms in terms of average execution cost, achieving reductions of 19%, 17.3%, and 21.5%, respectively. [ABSTRACT FROM AUTHOR]

Software-defined networking (SDN) is a network architecture where the control and data plane are separated. As the network size grows, relying on just one controller can lead to various problems. Thus, in highly scalable networks, multiple controllers are needed. This critical issue of determining the number and placement of controllers is known as the controller placement problem (CPP). In this paper, game theory is used to solve CPP by identifying the optimal number of controllers. Two algorithms, golden eagle optimization (GEO) and grey wolf optimization (GWO), are utilized to find the most efficient mapping between switches and controllers. Since CPP is a discrete problem, GEO and GWO have first been discretized and then hybridized to form a new algorithm called GEWO. This algorithm is used to discover the most efficient mapping between switches and controllers. Additionally, simulated annealing is employed for better local search. The effectiveness of this approach is evaluated using different numbers of controllers on four well-known software-defined networks from the Internet Topology Zoo. The results are compared against various existing algorithms in the field, and it is observed that GEWO outperforms the competition. The findings demonstrate that GEWO reduces load imbalance by 24.07%, decreases end-to-end delay by 20.95%, and lowers average energy consumption by 11.65%. [ABSTRACT FROM AUTHOR]

Distribution grid network has problems such as long mileage, large scale, complex surrounding environment, and aging of equipment. It is the development trend of power distribution network operation and maintenance to use unmanned aerial vehicles to patrol and combine with image processing technology for intelligent detection of equipment status. Image segmentation is well-known technique for extracting defect regions of equipment from distribution network inspection images. Therefore, this paper proposes an efficient a novel multilevel thresholding segmentation method to improve the fault diagnosis process with an improved weighted mean of vectors optimization (IINFO) algorithm. The IINFO algorithm adopts various measures to improve the optimization results, including Gaussian mutation to increase the local search ability and range of the optimal individual, Cauchy mutation to enhance the global search ability of its vector individual, reflective learning operators to strengthen self-learning and avoid local optimal solutions, and parallel operation to improve the utilization of computational resources. Moreover, two-dimensional Kapur cross-entropy is used as an objective function to solve the multilevel thresholding problem. The proposed method is evaluated using benchmark functions and distribution network inspection image datasets and is compared with 12 other metaheuristic algorithms. The results demonstrate that the proposed method has better performance and a higher ability to find optimal solutions compared to the other algorithms. These findings suggest that our method may be useful in improving the accuracy and efficiency of distribution network inspections and have significant potential for practical applications. [ABSTRACT FROM AUTHOR]

In this article, a novel population selection method, fitness distance balance (FDB), and predictive candidate (PC) solution generation hybridization with starling murmuration optimizer (SMO), FDBPC-SMO are proposed. In FDBPC-SMO algorithm, FDB selects subpopulations instead of the separating search strategy (SSS) in the original SMO. The separating size determined in SMO is given as input to the FDB, and the FDB generates the subpopulation based on the distances among the populations. The least squares strategy is applied to the population obtained at the end of the SMO, and the estimated population candidates are found and replaced with the worst solution candidates from the original population. By adding qualitative analysis, the effectiveness of the FDBPC-SMO has been examined based on the dimension and iteration. The success of FDBPC-SMO is the selection of more efficient candidate solutions from the previous population at each iteration, thus minimizing the possibility of getting stuck in the local optimum. The performance of FDBPC-SMO has been investigated on CEC2017 and CEC2019 test sets and seven engineering application problems. In addition, Wilcoxon and Friedman statistical tests confirm the convergence and fitness results of the proposed method. Accordingly, comparing to conventional and improved methods, it is clear that the convergence ability of FDBPC-SMO is superior. [ABSTRACT FROM AUTHOR]

Agriculture is one of the most crucial aspects of a nation's growth. However, the quality and quantity of crop yield are severely affected by various plant diseases. Plant diseases must be identified and prevented at an early stage to improve food quality and production rate. The emerging deep learning network of convolutional neural networks (CNNs) achieved excellent results in plant disease classification. However, the classification potential of the network depends largely on the configuration of hyperparameters. Finding the optimal set of hyperparameters is a tedious, time-consuming, and challenging task. To tackle such an issue, this paper proposes an optimized CNN integrated with a novel modified whale optimization algorithm (MWOA) to achieve plant disease classification. Here, the novel method of optimizing CNN hyperparameters facilitates quicker implementation. To boost the proposed model's efficiency, several data augmentation methods are used such as rotation, scaling, and flipping. The proposed MWOA-CNN model is implemented using a plant village database that includes fourteen plant species, 38 disease classes, and healthy leaves as well. The experiential findings showed that the proposed model outperforms the existing models by attaining the highest classification accuracy of 99.92%, resulting in an effective model for plant disease classification. [ABSTRACT FROM AUTHOR]

In the context of complex structures and blurred cell boundaries present in breast cancer histopathological tissue images under a microscope, traditional thresholding methods struggle to accurately separate lesion areas in breast cancer image segmentation. To address this challenge, we propose a multi-threshold segmentation method for breast cancer images based on an improved Dandelion Optimization algorithm. This approach incorporates the concept of opposite-based learning and utilizes the improved Dandelion Optimization algorithm to calculate the maximum between-class variance as the optimization objective. Moreover, the method establishes fallback strategies and incorporates a memory matrix, while leveraging the golden jackal energy judgment mechanism to identify optimal thresholds. The experimental results show that compared with the Crow search algorithm, Harris Hawks optimization algorithm, artificial gorilla troop optimization algorithm, dandelion optimization algorithm, ocean predator algorithm, whale optimization algorithm, sparrow search algorithm, and sine cosine algorithm, and the improved Dandelion optimization algorithm achieves the highest fitness value and converges at the fastest speed when using the same threshold number, it also occupies an advantageous position in terms of peak signal-to-noise ratio, structural similarity index, feature similarity index, and mean square error. [ABSTRACT FROM AUTHOR]

The influence maximization problem that has caused great attention in social network analysis aims at selecting a small set of influential spreaders so that the information cascade triggered by the seed set is maximized. The majority of the existing works mainly focus on developing single-stage seeding strategies that would ignite all the seeds before the influence spread. However, it cannot depict the scenarios of the practical, where ones would like to make further decisions based on observed activation. In this paper, we investigate the policies for the intractable sequential influence maximization problem. A Q-learning-driven discrete differential evolution algorithm based on the reinforcement Q-learning model, which is treated as a parameter controller to adaptively adjust the parameters during the evolution of the algorithm, is proposed. The policy distributes the seeding actions over the spreading process by estimating the latest node status of the network dynamically. Extensive simulations are conducted on six social networks of the practical, and the findings demonstrate the superiority and effectiveness of the hybrid meta-heuristic algorithm compared with the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

This paper presents a novel and powerful population-based metaheuristic algorithm called Nizar Optimization Algorithm (NOA). This algorithm is based on two techniques. The first technique is to use effective mappings, which are divided into two types: mixing and transformation mappings. The mixing mappings return a mixed vector by replacing or shuffling the elements of two selected vectors, and the transformation mappings return the translation, dilation, or transfer of a selected vector. The second technique is to determine the effective points by using the effective mappings and individuals of the population, and then, these points are used in the learning process of NOA. To validate and demonstrate the performance of the proposed algorithm and its ability to balance exploration and exploitation, NOA is tested on 60 unconstrained benchmark functions and four classic constrained real-world engineering problems. The experimental results are verified by a comparative study with over 20 well-known and recently developed optimization algorithms. These results show that the proposed algorithm outperforms all other algorithms in terms of solution accuracy, convergence curve speed, statistical measurements, and computational expenses. [ABSTRACT FROM AUTHOR]

Optimizing resource utilization and train scheduling is essential to satisfy passengers and reduce operating costs. This study develops the train schedule under scenario-oriented stochastic conditions. The proposed approach is a multi-objective mathematical-based mixed integer linear programming (MILP) approach; the objective is to minimize the average passenger expectation and the total number of train operation cycles. The non-dominated sorting genetic algorithm (NSGA-II) has been developed with multi-crossover and multi-mutation operators, then hybrid with simulating annealing (SA) operator (NSGA-II-SA). The model with four meta-heuristic algorithms has been technically analyzed. In a case study, the train schedule in the double-track rail network of the Tehran–Mashhad railway of Iran has been compared with the golden point. Experimental results show that a proposed approach can suitably fit the problem considering important metrics with an improvement of %7.34 and %6.89 for the average passenger waiting time and the total number of train operation cycles, respectively. [ABSTRACT FROM AUTHOR]

This paper proposes a robust optimization of eight chaotic maps: Logistics, Sine, Gauss, Circle, Tent, Chebyshev, Singer, and Piecewise Maps, for superior image encryption. The proposed model consists of two main processes: chaotic confusion and pixel diffusion. In the chaotic confusion process, the positions of the image's pixels are permuted with the chaotic maps, where the initial condition and the control parameters represent the confusion key. Firstly, the confusion process was performed using the eight chaotic maps without optimization. Then nine metaheuristic optimizers, which are the genetic algorithm, particle swarm optimizations, whale optimization algorithm, dragonfly algorithm, grey wolf optimizer, moth-flame optimizer, sine cosine algorithm, multi-verse optimizer, and ant-lion optimizer, have been used to fine-tune the control parameters of the eight chaotic maps. Then the image's pixel values are changed using the diffusion function in the pixel diffusion process. Multiple performance metrics, such as entropy, histogram, cross-correlation, computation time analysis, the number of pixels change rate (NPCR), unified average changing intensity (UACI), noise attack, data loss, and key analysis metrics, are utilized to evaluate the proposed model. The results demonstrate that the encryption algorithms based on the eight optimized chaotic maps are more resistant to differential attacks than those without optimization. Furthermore, the optimized Gauss chaotic map is the most computationally efficient, while the chaotic circle map has the most robust key. The careful adjustment of initial conditions and control parameters empowers the chaotic maps to create encryption keys with greater randomness and complexity, thereby increasing the security level of the encryption scheme. Experimental analysis indicates that the correlation coefficient values of images encrypted with the proposed scheme are nearly zero, the histogram of the encrypted images is uniform, the execution time of 0.1556 s, the key space of 10^80, NPCR of 99.63%, UACI of 32.92%, and entropy of 7.997. Moreover, the analysis of noise and cropping attacks, along with the comparison with other algorithms, demonstrate the efficiency and robustness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

The issue of power system protection remains a crucial aspect for power engineers. One of the important protection devices in distribution networks is the overcurrent relay, which requires constant attention to ensure its performance is optimized. This paper applied very first time the Archimedes principle-based Archimedes optimization algorithm (AOA) on an optimal overcurrent relay coordination problem from the power system protection area. The proposed algorithm is evaluated on two unique case studies, the IEC microgrid benchmark and a combination of overhead/underground cable systems, to verify its efficiency. The uniqueness of these case studies is that in microgrids due to the incorporation of wind turbines bidirectional current flows whereas in overhead/underground system magnitude of current will be different in the various part of the network. Because of this impact of current in both networks, the relay coordination job becomes challenging. The results demonstrate that the proposed AOA technique outperforms other existing algorithms, such as Particle Swarm Optimization, Water Cycle Algorithm (WCA), Whale Optimization Algorithm, Crow Search Algorithm, and Evaporation Rate WCA, in terms of computational efforts and performance. Furthermore, the proposed algorithm effectively minimizes the total operating time for both networks without violating any constraints. The study confirms that the proposed algorithm provides an efficient solution for the overcurrent relay coordination problem, which can be beneficial for power system protection. The Archimedes principle-based Archimedes optimization algorithm is demonstrated to be a promising optimization tool that can effectively solve similar optimization problems in the future. [ABSTRACT FROM AUTHOR]

Sentiment Analysis (SA) is a domain- or topic-dependent task since polarity terms convey different sentiments in various domains. Hence, machine learning models trained on a specific domain cannot be employed in other domains, and existing domain-independent lexicons cannot correctly recognize the polarity of domain-specific polarity terms. Conventional approaches of Topic Sentiment Analysis perform Topic Modeling (TM) and SA sequentially, utilizing the previously trained models on irrelevant datasets for classifying sentiments that cannot provide acceptable accuracy. However, some researchers perform TM and SA simultaneously using topic-sentiment joint models, which require a list of seeds and their sentiments from widely used domain-independent lexicons. As a result, these methods cannot find the polarity of domain-specific terms correctly. This paper proposes a novel supervised hybrid TSA approach, called Embedding Topic Sentiment Analysis using Deep Neural Networks (ETSANet), that extracts the semantic relationships between the hidden topics and the training dataset using Semantically Topic-Related Documents Finder (STRDF). STRDF discovers those training documents in the same context as the topic based on the semantic relationships between the Semantic Topic Vector, a newly introduced concept that encompasses the semantic aspects of a topic, and the training dataset. Then, a hybrid CNN–GRU model is trained by these semantically topic-related documents. Moreover, a hybrid metaheuristic method utilizing Grey Wolf Optimization and Whale Optimization Algorithm is employed to fine-tune the hyperparameters of the CNN–GRU network. The evaluation results demonstrate that ETSANet increases the accuracy of the state-of-the-art methods by 1.92%. [ABSTRACT FROM AUTHOR]

To address the problems of slow convergence of the traditional whale optimization algorithm and the tendency to fall into local optimal solutions, an improved algorithm based on a combination of elite disturbance opposition-based learning and dynamic spiral updating (OWOA) is proposed, which improves the convergence speed of the algorithm and increases the probability of the algorithm jumping out of the local optimum to find the global optimum. First, an elite disturbance opposition-based learning strategy is used to initialize the whale population to ensure diversity of the population, and then, random perturbation of the elite whales increases the exploratory capacity of the population to avoid falling into local optimal solutions. Secondly, a dynamic spiral update strategy is used to dynamically adjust the shape of the spiral with the number of iterations, avoiding the algorithm from falling into a local optimum at the beginning of the iteration and improving the convergence speed of the algorithm at a later stage, which improves the ability of the algorithm to find the global optimum. The experimental results show that the OWOA algorithm, due to the introduction of the lite disturbance opposition-based learning strategy and dynamic spiral updating strategy, has higher convergence speed and accuracy on single-peak functions, multi-peak functions and multi-dimensional functions compared to other algorithms. Furthermore, good results have been obtained in the practical application of OPF. This shows that our improvements are effective. [ABSTRACT FROM AUTHOR]

The Mayfly Algorithm (MA) is a widely used metaheuristic algorithm characterized by a simple structure with simple parameters. However, MA may have problems such as poor global search ability and tend to fall into local optima. To overcome these limitations, this paper presents a chaos-based mayfly algorithm with opposition-based learning and Levy flight (COLMA) to boost the global search and local exploitation performance. In COLMA, we first introduced tent chaos to optimize the initialization process of the mayfly population, as random initialization processes may result in low diversity of the mayfly population. In addition, the gravity coefficient has been replaced with an adaptive gravity coefficient to balance the global search ability and local exploitation ability during the iterative process of the algorithm. In the process of updating the position of the male mayfly population, an opposition-based learning strategy based on an iterative chaotic map with infinite collapses is adopted to prevent the male mayfly population from falling into local optima. At the same time, in order to solve the problem of small search range of female mayfly population, Levy flight strategy was introduced to replace random walk strategy. Finally, an offspring optimization strategy was proposed to increase the probability of the offspring mayfly population approaching the optimal solution. To verify the effectiveness and superiority of COLMA and the adopted strategy, experiments were conducted on the classical benchmark functions, CEC 2017 benchmark suite and CEC 2020 real-world constraint optimization problems, and the results were statistically tested using the Wilcoxon signed rank test and Friedman test. The analysis results show that the proposed COLMA has statistical validity and reliability and has great advantages compared with MA, variant MA in terms of optimization accuracy, stability and convergence speed. [ABSTRACT FROM AUTHOR]

The Internet of Robotic Things (IoRT) is an integration between autonomous robots and the Internet of Things (IoT) based on smart connectivity. It's critical to have intelligent connectivity and excellent communication for IoRT integration with digital platforms in order to maintain real-time engagement based on efficient consumer power in new-generation IoRT apps. The proposed model will be utilized to determine the optimal way of task offloading for IoRT devices for reducing the amount of energy consumed in IoRT environment and achieving the task deadline constraints. The approach is implemented based on fog computing to reduce the communication overhead between edge devices and the cloud. To validate the efficacy of the proposed schema, an extensive statistical simulation was conducted and compared to other related works. The proposed schema is evaluated against the Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Whale Optimization Algorithm (WOA), Artificial Bee Colony (ABC), Ant Lion Optimizer (ALO), Grey Wolf Optimizer (GWO), and Salp Swarm Algorithm to confirm its effectiveness. After 200 iterations, our proposed schema was found to be the most effective in reducing energy, achieving a reduction of 22.85%. This was followed closely by GA and ABC, which achieved reductions of 21.5%. ALO, WOA, PSO, and GWO were found to be less effective, achieving energy reductions of 19.94%, 17.21%, 16.35%, and 11.71%, respectively. The current analytical results prove the effectiveness of the suggested energy consumption optimization strategy. The experimental findings demonstrate that the suggested schema reduces the energy consumption of task requests more effectively than the current technological advances. [ABSTRACT FROM AUTHOR]

Many activities are conducted on social media platforms, such as promoting products, sharing news and sharing achievements. As a result of users' freedom and anonymity on social media platforms, hate speech and harassment are common. Posts that spread hate and offense should be detected and deleted as soon as possible as they spread very quickly and have many negative consequences for the human race. The detection task becomes significantly more difficult when online users use code-mixed text in places where English is not the primary language. The problem of hate speech detection has recently been reduced to a binary classification task, without taking into account its topical focus and its target-oriented nature. Because there is no combined annotated dataset and scientific study that can provide insight into the relationship between offense traits, existing techniques usually only examine one or two offense traits at a time. Furthermore, these techniques are not efficient for multilingual, where most conversations are code-mixed. In this paper, we propose an optimal feature extraction and hybrid diagonal gated recurrent neural network (FE-DGRNN) for hate speech detection and sentiment analysis in multilingual code-mixed texts. The proposed FE-DGRNN technique consists threefold processes. After preprocessing, we first introduce an improved seagull optimization (ISO) algorithm for multiple features extraction from given code-mixed texts. Then, we utilize a quantum search optimization algorithm to optimize the extracted features which reduces the data dimensionality issues in further detection phase. Next, a hybrid diagonal gated recurrent neural network (Hyb-DGRNN) introduces to detect hate speech and analyzes sentiment on their language. In order to validate the effectiveness of our proposed FE-DGRNN technique, we conducted experiments on the HASOC 2019 dataset. This dataset includes posts written in English, Hindi and German, allowing us to evaluate the performance of our approach across multiple languages. From the simulation results, we observed that the accuracy of FE-DGRNN is 87.74%, 88.98% and 84.74% for Task-1, Task-2 and Task-3, respectively, for multilingual code-mixed texts dataset. Overall, the proposed FE-DGRNN technique shows a significant improvement in accuracy, precision, recall and F-measure compared to other classifiers, indicating its potential to be a robust and effective tool for hate speech detection and sentimental analysis in multilingual code-mixed texts. [ABSTRACT FROM AUTHOR]

Fog computing paradigm attempts to provide diverse processing at the edge of IoT networks. Energy usage being one of the important elements that may have a direct influence on the performance of fog environment. Effective scheduling systems, in which activities are mapped on the greatest feasible resources to meet various competing priorities, can reduce energy use. Consequently, a hybrid discrete optimization method called HDSOS-GOA, which uses the Dynamic voltage and frequency scaling (DVFS) approach, is proposed to handle scientific workflow scheduling challenges in the fog computing environment. HDSOS-GOA combines the search qualities of Symbiotic Organisms Search (SOS) and the Grasshopper Optimization Algorithm (GOA) algorithms and the selection of these algorithms for performing workflow scheduling is based on the probability calculated by the learning automata. The HEFT method is used to determine the task sequence. Our solution focuses on reducing the energy consumption of the scheduling process by reducing the number of Virtual Machines required for workflow execution in addition to optimizing the makespan. Comprehensive experiments are carried out on four different scientific workflows with different sizes with and without deadline constraints to evaluate the performance of the suggested scheduling strategy. The results of the experiments show that scheduling with the suggested approach outperforms other well-known metaheuristic algorithms. [ABSTRACT FROM AUTHOR]

Minimum cross-entropy is widely used to find the best threshold values for image segmentation; this technique is known as MCET. However, when the number of thresholds increases, it becomes computationally expensive. Under such circumstances, employing a metaheuristic algorithm (MA) is a good choice. The Runge Kutta (RUN) optimization algorithm is MA recently proposed for solving global optimization problems. Like other MA, the RUN tends to fall into sub-optimal solutions presenting a premature convergence, especially in high-dimensional problems. This article aims to improve the RUN by merging it with opposition-based learning (OBL), creating a hybrid algorithm called RUN-OBL. By doing this hybridization, the RUN can search in two directions, and its performance is considerably improved, solving its drawbacks. One of the most significant improvements in this paper is that the RUN-OBL can handle high-dimensional search spaces escaping from local optimal solutions. The performance of the RUN-OBL is tested over different experiment series. Firstly, the RUN-OBL is tested using the CEC'17 set of benchmark functions for numerical optimization. Here, comparisons with other MA and the Friedman test were conducted. Secondly, the proposed algorithm segments digital images using the MCET. A set of benchmark images with different degrees of complexity are used. The RUN-OBL for image thresholding is tested in the third experiment series over a set of medical images (chest x-ray). A comparative study of the segmentation results is conducted to verify the efficiency of the proposal. Here different MA and other image segmentation methodologies were used. The comparisons were performed in terms of the fitness functions, peak signal-to-noise ratio (PSNR), feature similarity index (FSIM), and structural similarity index (SSIM). Experimental results demonstrate that the proposed RUB-OBL approach is better regarding quality and consistency for segmentation purposes and the optimization of complex problems. [ABSTRACT FROM AUTHOR]

The Harris hawks optimization (HHO) algorithm is a new metaheuristic algorithm proposed in recent years. Due to the shortcomings of this algorithm in solving complex high-dimensional optimization problems with a slow convergence speed, low accuracy, and the high likelihood to fall into local optimum, a mixed Harris hawks optimization (MHHO) algorithm based on the pinhole imaging strategy is proposed, including four strategies to improve the optimization performance. Firstly, the pinhole imaging strategy is used to enable the Harris' hawks to approach the optimal solution faster and accelerate convergence. Secondly, the spiral parameter is introduced into the exploration phase to help the searching paths of the Harris' hawks more diverse and improve the global search ability of the algorithm. Finally, the greedy strategy of the aquila optimization algorithm and the position update strategy of the flower pollination optimization algorithm are embedded in the exploitation stage to make the algorithm jump out of local optimum effectively. To verify the effectiveness of the proposed MHHO algorithm, it is compared with the classical HHO algorithm and 16 other state-of-the-art algorithms, and extensively tested on 23 well-known benchmark functions, the IEEE CEC2017 test sets, and three complex constrained engineering optimization problems. The test results show that MHHO achieves the top ranking on both the benchmark functions and the CEC2017 test sets, demonstrating its superior performance in terms of faster convergence speed and higher accuracy. [ABSTRACT FROM AUTHOR]

This paper presents a hybrid simplified grey wolf optimization algorithm combining with an modified pigeon inspired optimization algorithm. Firstly, a method of population initialization in the restricted area is designed to improve the average fitness of individuals. Then, a decay factor is designed to improve the search accuracy. At the same time, a new search position update formula is designed to balance the global search and local search ability, so as to improve the convergence speed of the algorithm. Furthermore, in order to verify the performance, the convergence, complexity and accuracy of the algorithm are analyzed by using linear difference equations. Finally, the comparison between different meta-heuristic optimization algorithms, the influence of parameters and their application in the path planning of unmanned aerial vehicle (UAV) are tested. Simulation results show that the proposed algorithm has stronger optimization ability and better robustness than those of compared algorithms. [ABSTRACT FROM AUTHOR]

In many fields, it is a common practice to collect large amounts of data characterized by a high number of features. These datasets are at the core of modern applications of supervised machine learning, where the goal is to create an automatic classifier for newly presented data. However, it is well known that the presence of irrelevant features in a dataset can make the learning phase harder and, most importantly, can lead to suboptimal classifiers. Consequently, it is becoming increasingly important to be able to select the right subset of features. Traditionally, optimization metaheuristics have been used with success in the task of feature selection. However, many of the approaches presented in the literature are not applicable to datasets with thousands of features because of the poor scalability of optimization algorithms. In this article, we address the problem using a cooperative coevolutionary approach based on differential evolution. In the proposed algorithm, parallelized for execution on shared-memory architectures, a suitable strategy for reducing the dimensionality of the search space and adjusting the population size during the optimization results in significant performance improvements. A numerical investigation on some high-dimensional and medium-dimensional datasets shows that, in most cases, the proposed approach can achieve higher classification performance than other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The economic emission dispatch problem is one of the most crucial problems for the future smart power network. The integration of renewable energy resources and the use of electric vehicles (EVs) are various techniques that have been suggested to deal with the increase in greenhouse gas emissions. However, to carefully examine various scenarios of integration of the above elements is important to reduce emission rates while considering economic constraints. In this research, the effectiveness of the use of photovoltaics (PVs) and EVs' discharging/charging behavior is investigated from various perspectives in a hybrid economic emission dispatch. Here, a multi-objective model includes the optimization of two objectives: cost and pollution is suggested. Based on a penalty factor, we convert the multi-objective function into a single objective. For solving this non-convex, nonlinear and on-smooth problem, a modified whale optimization algorithm (WOA) is exerted to gain the optimum results. The idea of two sub-modifications is derived from Lévy flight algorithm and the mutation operator of the genetic algorithm. The verification of the suggested model is performed by the IEEE 30-bus system with 6-units. The simulated results illustrate that the incorporation of PV along with EVs in economic dispatch problem can improve the pollution emission, the economic costs, and convergence rate, effectively. [ABSTRACT FROM AUTHOR]

Parameter identification for complex systems of nonlinear nature is challenging due to the complicated process structure and large number of parameters with different identifiability. The scope of this work is to develop a model-based parameter identification method for a nonlinear industrial winding system. The proposed parameter identification method consists of two key steps: First, enhanced whale optimization algorithm (EWOA) was proposed to alleviate the issues of low search performance and premature convergence of WOA, for which the following enhancements were made to WOA: (1) improvements to the bubble-net strategy of its mathematical model, (2) amendments to the humpback whales' movements in the direction of the best whales, and (3) imitation of the schooling behavior of humpback whales when chasing prey. Second, EWOA was acted as a training method for artificial neural networks (ANNs)-type multilayer perceptron (MLP), a method referred to as EWOA-MLP. In this, EWOA was applied to train ANNs in order to mitigate the main difficulties of ANNs due to their nonlinear nature and unknown optimal set of control parameters (i.e., weights and biases). The performance of the proposed EWOA-MLP was assessed in modeling the subsystems of the winding process under study and in solving fifteen classification datasets. The effectiveness of EWOA-MLP in both modeling and classification studies was judged by several pertinent assessment metrics. Results of comparison of the proposed EWOA-MLP with other promising methods firmly confirm the promising performance of EWOA-MLP for both local optima avoidance and convergence rate, proving its value and superiority. Moreover, EWOA-MLP was able to outperform other algorithms in modeling the winding process and solving many classification problems. [ABSTRACT FROM AUTHOR]

Block size in a permissioned blockchain system plays a huge role in scalability and performance of system. A large number of blocks are continuously disseminated into the network and hence it becomes important to study the effect of block size in a decentralized peer-to-peer network. The scalability of the system is the biggest concern. Scalability is a metric that refers to the ability of the system to increase or balance its performance depending upon the load and the processing demands on the system. The success of any blockchain-based application is dependent on the size of block used. Block size governs the time required to transmit a block (block transmission time) and the rate at which the unconfirmed transactions are verified (transaction pool clearance time). Block size also acts as a factor in ordering the blocks in the chain and dictates the performance and security of the system. This paper presents simulation analysis of the influence of block size on blockchain and the role of network bandwidth-aware optimal block size on the performance of the blockchain system. Nature-inspired algorithms have been used to determine the link between block size and scalability and are used for finding the optimal block size for a Blockchain-based application. [ABSTRACT FROM AUTHOR]

A lot of research studies focus on the development of a new algorithm or the techniques which improve the performance of the original algorithm. Very few studies conduct the research on the effect of the initial population on the solution quality of algorithms. However, in these studies, one or two algorithms have been used, and a limited number of problems have been handled. To fill in the gap in the literature, this study presents a comprehensive analysis of the five algorithms on the effect of the initial population on their final results including both the numerical and real-world problems along with a wide variety of types of distributions. The study consisted of three rounds and followed the strategy for determining the candidate algorithms to be participated in the next rounds, supported by the statistical tests. Rather than using popular random numbers, fourteen different distributions are used to imitate the random numbers in the initial population generation mechanisms of the algorithms. Two different numerical benchmark sets along with nine real-world problems are used to evaluate the performance of the algorithms. The results are compared with the original ones and other distribution-integrated algorithms. Since knowledge of the appropriate random number source is not available a priori, this study could be a good foundation for future studies not only on the matter of the effect of several distributions on the performances of the algorithms but also introducing an alternative way in generating an initial population. [ABSTRACT FROM AUTHOR]

In this study, we design, evaluate, and compare multiple heuristic techniques for mission scheduling of distributed systems comprising unmanned aerial vehicles (UAVs) in energy-constrained dynamic environments. These techniques find effective mission schedules in real-time to determine which UAVs and sensors are used to surveil which targets. We develop a surveillance value metric to quantify the effectiveness of mission schedules, incorporating the amount and usefulness of information obtained from surveilling targets. We use the surveillance value metric in simulation studies to evaluate the heuristic techniques with a reality-based randomized model. We consider two comparison heuristics, three value-based heuristics, and a metaheuristic that intelligently switches between the best value-based heuristics. Additionally, preemption and filtering techniques are applied to further improve the metaheuristic. We show that, for all scenarios that we consider, the novel modified metaheuristics find solutions that are the best on average compared to all other techniques that we evaluate. [ABSTRACT FROM AUTHOR]

Feature selection identifies the most relevant features that help improve the classifier's performance. The Binary Multi-Objective Grey Wolf Optimizer-Sigmoid (BMOGWO-S) algorithm was recently introduced for feature selection in classification, which is a multi-objective approach to feature selection. It was identified to be providing better results than other multi-objective feature selection algorithms that are currently available. It is a metaheuristic algorithm that implements a wrapper method of feature selection. Usually, there are many statistical methods that are used for finding a relation between the input and output variables in a dataset. The BMOGWO-S algorithm does not utilize any statistical information from the dataset and relies entirely on metaheuristics. In this research, a Statistically aided Binary Multi-Objective Grey Wolf Optimizer-Sigmoid (SaBMOGWO-S) is proposed, which uses the advantages of statistical information about the dataset's attributes. Also, methods are introduced to reduce the algorithm's running time by avoiding unnecessary computations. Results of the proposed algorithm are compared against the results obtained from the existing state-of-the-art methods with respect to 21 standard datasets from the UCI repository. For higher dimensional datasets with more than 100 features, the proposed algorithm has found to be outperforming the other methods in terms of reduction in features and classification error rate and for the lower dimensional datasets it has outperformed others in terms of run time. The best error rate obtained is 0.00 for some datasets, and the average error rate obtained for all datasets with the proposed SaBMOGWO-S is 0.11. [ABSTRACT FROM AUTHOR]

The growth of the Internet of Things (IoT) can lead to improved productivity, scalability, connectivity, and saving time and money. However, the increasing number of IoT-based applications has MAPE the centralized cloud computing paradigm face several challenges such as latency and bandwidth. Nowadays, fog computing with distributed architecture has emerged to support latency-sensitive IoT applications with limited resources. Using the fog instead of the cloud can bring the storage and computing facilities closer to the edge of the network and thus provide better performance for the end user. Meanwhile, IoT applications usually have complex multi-component structures whose efficient placement on fog nodes can overcome the resource restrictions of IoT devices. This problem as the IoT Service Placement Problem (SPP) is NP-hard, where nature-inspired approaches widely provide robust solutions to solve it. In this paper, a meta-heuristic-based evolutionary approach named SPP-ICA is presented to address SPP, which originates from the Imperialist Competitive Algorithm (ICA). SPP-ICA is developed based on MAPE-K autonomous planning model, so that it applies service deployment priority and resource consumption distribution in the placement process. ICA embedded in SPP-ICA leads to more effective placement of IoT services in terms of latency and resource utilization by considering the concepts of elitism and balanced resource consumption. Experimental results show that SPP-ICA performs significantly better than state-of-the-art algorithms with meta-heuristic structure. On average, SPP-ICA deploys more IoT services on fog and reduces service latency by 9%. [ABSTRACT FROM AUTHOR]

Microarray data have a crucial role in identifying and classifying different types of cancer tissues. In cancer research, high dimensional of microarray datasets has always caused problems in the design of classifiers. Thus, microarray data before classification are preprocessed through feature selection (FS) techniques, whereby the features with lower information values are discarded away. Traditional FS as search involves selecting a subset of features from large number of features. On the other hand, other subsets of features also contain information, and the multimodal property of FS problem should be considered. In multimodal metaheuristic searches, instead of one solution, several semi-optimized solutions are found, where each solution is an independent view compared to others. In this paper, in the first stage, multimodal FS is performed using multimodal firefly algorithm (FA), where several feature subsets are obtained. Then, in the second stage, this subset of features is used for assigning weight to the classifiers and identifying the effective classifiers in the ensemble process based on Dempster–Shafer evidence theory. The mentioned weights are calculated so that an ensemble could be developed through weighted majority vote. Eventually, the proposed algorithm is evaluated on microarray data for cancer diagnosis. The proposed method efficiency is evaluated by applying on 11 microarray datasets. The obtained results indicated good performance of the multimodal FS method in relation to other methods compared. [ABSTRACT FROM AUTHOR]

This study aims to use a machine learning (ML)-based enhanced diagnosis and survival model to predict heart disease and survival in heart failure by combining the cuckoo search (CS), flower pollination algorithm (FPA), whale optimization algorithm (WOA), and Harris hawks optimization (HHO) algorithms, which are meta-heuristic feature selection algorithms. To achieve this, experiments are conducted on the Cleveland heart disease dataset and the heart failure dataset collected from the Faisalabad Institute of Cardiology published at UCI. CS, FPA, WOA, and HHO algorithms for feature selection are applied for different population sizes and are realized based on the best fitness values. For the original dataset of heart disease, the maximum prediction F-score of 88% is obtained using K-nearest neighbour (KNN) when compared to logistic regression (LR), support vector machine (SVM), Gaussian Naive Bayes (GNB), and random forest (RF). With the proposed approach, the heart disease prediction F-score of 99.72% is obtained using KNN for population sizes 60 with FPA by selecting eight features. For the original dataset of heart failure, the maximum prediction F-score of 70% is obtained using LR and RF compared to SVM, GNB, and KNN. With the proposed approach, the heart failure prediction F-score of 97.45% is obtained using KNN for population sizes 10 with HHO by selecting five features. Experimental findings show that the applied meta-heuristic algorithms with ML algorithms significantly improve prediction performances compared to performances obtained from the original datasets. The motivation of this paper is to select the most critical and informative feature subset through meta-heuristic algorithms to improve classification accuracy. [ABSTRACT FROM AUTHOR]