Search

Data analysis becomes more important since rapid technology developments. In data analysis, data clustering is one of the very useful approaches. It can reveal important information hiding inside the dataset by organizing the instances based on their similarity. The objectives of data clustering are maximizing dissimilarity between clusters and minimizing dissimilarity within clusters. In order to construct a good clustering results, many clustering algorithms have been proposed, including the metaheuristic-based clustering algorithms. Recently, a new metaheuristic algorithm named gradient evolution has been proposed. This algorithm shows a good performance on solving the optimization problems. Therefore, this paper employs this GE algorithm for solving the clustering problem. In order to obtain a better clustering result, this paper considers multi-objective clustering instead of single-objective clustering. In this paper, the original GE algorithm is improved so then it is suitable for the multi-objective problem. The proposed modification includes the procedure for vector updating and jumping which involves Pareto rank assignment. In addition, it also employs K-means algorithm to provide the final clustering result. The proposed algorithm is verified using some benchmark datasets. It is also compared with some other multi-objective metaheuristic-based clustering algorithms. The experimental results show that the proposed algorithm can obtain better results than other metaheuristic-based algorithms. [ABSTRACT FROM AUTHOR]

Inventory control is one of the most critical issues in corporate management. Many mathematical models have been developed to optimize control strategies for the companies' inventory. Economic production quantity (EPQ) is one of the classic models for inventory control, which is widely used. To deal with the uncertainty in the real world, we need to develop new and useful models for modeling systems of inventory management. In such cases, fuzzy models play a unique role in the field of inventory management. The main contribution of this study is to apply some well-known metaheuristic to solve an extended EPQ model based on fuzzy numbers considering multiple deliveries. This study aims to develop an EPQ model by considering demand as triangular fuzzy numbers and multiple deliveries (delivering in multiple packages) and by considering limitations in warehouse space as well as the total number of orders. Given these conditions, EPQ costs are calculated, and new modeling is presented. The obtained fuzzy model has been simplified by using the α -cut and changing the variables, and finally, the most well-known metaheuristic algorithms, GA, PSO, GWO, and ICA, are applied in different problem sizes, and obtained results are analyzed in terms of minimizing cost function and CPU time. The result of this paper shows that GWO has superior performance in terms of various parameters. [ABSTRACT FROM AUTHOR]

This paper presents a novel metaheuristic algorithm named as life choice-based optimizer (LCBO) developed on the typical decision-making ability of humans to attain their goals while learning from fellow members. LCBO is investigated on 29 popular benchmark functions which included six CEC-2005 functions, and its performance has been benchmarked against seven optimization techniques including recent ones. Further, different abilities of LCBO optimization algorithm such as exploitation, exploration and local minima avoidance were also investigated and have been reported. In addition to this, scalability is tested for several benchmark functions where dimensions have been varied till 200. Furthermore, two engineering optimization benchmark problems, namely pressure vessel design and cantilever beam design, were also optimized using LCBO and the results have been compared with recently reported other algorithms. The obtained comparative results in all the above-mentioned experimentations revealed the clear superiority of LCBO over the other considered metaheuristic optimization algorithms. Therefore, based on the presented investigations, it is concluded that LCBO is a potential optimizer for engineering problems. [ABSTRACT FROM AUTHOR]

The natural aggregation algorithm (NAA) is a new efficient population-based optimizer. The NAA has a competent performance when compared to other well-established optimizers. However, a problem of concern is NAA lack of exploitation in its local search. In this article, we propose an improved version of NAA. The modifications made are: hypercubes with displacement and shrink mechanism applied in each shelter, we designed a new movement operator to search inside the hypercubes, an improved readjustment of the algorithm's parameters and "leave shelter" formula of NAA, to better mimic the aggregation behavior. To prove the effectiveness of the modified hypercube natural aggregation algorithm (HYNAA), we compared with classics optimizers, such as PSO, DE and ABC, state of the art, such as CMA-ES, MSA and NAA himself with a benchmark of 28 functions. The said functions consist of five unimodal, 19 multimodal and four hybrids, and we compared them on 30, 50 and 100 dimensions. We also made extra comparisons against NAA in 500 and 1000 dimensions to contrast the ability of the hypercubes to reduce the dimensional complexity. Finally, we tested two trajectory optimization problems. Experimental results and statistical tests demonstrate that the performance of HYNAA is significantly better than that of other optimizers. [ABSTRACT FROM AUTHOR]

This paper introduces a new optimization algorithm called electron radar search algorithm (ERSA) inspired by the electron discharge mechanism. It is based on the natural phenomenon of electric flow as the form of electron discharge through a gas, liquid, or solid environment. When the voltage between separated electrodes (anode and cathode) increases, electrons tendency to emission from a low potential state to the higher potential condition is grown up. However, electrons are trying to find the best path with the least resistance in the medium. At each point, electrons evaluate the surrounding environment with a radar mechanism and least resistance path is selected for the next move. Hence, in this paper, a novel developed meta-heuristic algorithm based on the electrons' search approach is presented and the algorithm is benchmarked on 20 mathematical functions with four well-known methods for validation and verification tests. Moreover, the algorithm is implemented in two engineering design problems (tension/expression spring and welded beam design optimization) and the results demonstrate that the ERSA performs more efficiently for solving unknown search spaces and the algorithm found best solution in approximately 95% of the reviewed benchmark functions. [ABSTRACT FROM AUTHOR]

Globalization has enlightened market with both opportunity and risk by bringing a more connected business environment, which ensures more customers and new markets. In contrast, it also brought a larger extent of competitors. The more collaborative environment can help companies to focus on their core competence to simultaneously reduce cost and participate more profitably in their trade. Uncertainty is a major outcome of the globalization process; firms are developing new methods and strategies to deal with risk and take control of uncertainty factors. This work introduces a novel approach that could help in collecting the end-of-life and end-of-use products from the end-users. These collected products enter the value chain and help in reducing the overall cost of the supply chain. A mixed-integer linear programming model has been formulated to assess the overall cost of the supply chain for the presented study. Due to the NP-hardness of the problem, few well-known metaheuristics and hybrid approaches are proposed as solution techniques for the first time. The Taguchi method is used to obtain the best combinations of algorithm parameters. In addition, problem instances are generated to validate the proposed model for a real-world case. Finally, the effectiveness of the algorithms is compared by using different criteria. [ABSTRACT FROM AUTHOR]

The building of a system for exploring the opinions of users that are made in the blog posts, tweets, reviews or comments regarding a particular topic, policy or a product is known as sentiment analysis. The primary aim of this is the determination of the user attitude regarding a certain topic. The harmony search algorithm has proved to be extremely useful in a varied range of problems in optimization. This shows better performance compared to the other techniques of optimization. Another very powerful technique that is applied to machine learning which is now getting extremely popular is gradient boosting. There are several tree parameters which have been optimized for the random forest and the gradient boosting machine that make use of the harmony search algorithm. [ABSTRACT FROM AUTHOR]

Various metaheuristic optimization algorithms are being developed to obtain optimal solutions to real-world problems. Metaheuristic algorithms are inspired by various metaphors, resulting in different search mechanisms, operators, and parameters, and thus algorithm-specific strengths and weaknesses. Newly developed algorithms are generally tested using benchmark problems. However, for existing traditional benchmark problems, it is difficult for users to freely modify the characteristics of a problem. Thus, their shapes and sizes are limited, which is a disadvantage. In this study, a modified Gaussian fitness landscape generator is proposed based on a probability density function, to make up for the disadvantages of traditional benchmark problems. The fitness landscape developed in this study contains a total of six features and can be employed to easily create various problems depending on user needs, which is an important advantage. It is applied to quantitatively evaluate the performance and reliability of eight reported metaheuristic algorithms. In addition, a sensitivity analysis is performed on the population size for population-based algorithms. Furthermore, improved versions of the metaheuristic algorithm are considered, to investigate which performance aspects are enhanced by applying the same fitness landscape. The modified Gaussian fitness landscape generator can be employed to compare the performances of existing optimization algorithms and to evaluate the performances of newly developed algorithms. In addition, it can be employed to develop methods of improving algorithms by evaluating their strengths and weaknesses. [ABSTRACT FROM AUTHOR]

In the current study, the performance of three evolutionary algorithms, differential algorithm (DE), evolution strategy (ES), and biogeography-based optimization algorithm (BBO), is examined for foundation design optimization. Moreover, four recent variations of evolutionary-based algorithms [i.e., improved differential evolution algorithm based on an adaptive mutation scheme, weighted differential evolution algorithm (WDE), linear population size reduction success-history-based adaptive differential evolution algorithm, and biogeography-based optimization with covariance matrix-based migration] have been tackled for handling the current problem. The objective function is based on the cost of shallow foundation designs that satisfy ACI 318-05 requirements is formulated as the objective function. This study addresses shallow footing optimization with two attitudes, routine optimization, and sensitivity analysis. As a further study, the effect of the location of the column at the top of the foundation is examined by adding two additional design variables. Three numerical case studies are used for both routine and sensitivity analysis. Moreover, the most common evolutionary-based technique, genetic algorithm (GA), is considered as a benchmark to evaluate the proposed methods' efficiency. Based on the results, there is no algorithm which works as the most efficient solver over all the cases; while, BBO and WDE showed an acceptable performance because of satisfying records in most cases. There were several cases in which GA, DE, and ES were incapable of finding a valid solution which meets all the constraints simultaneously. [ABSTRACT FROM AUTHOR]

In this paper, a new optimization algorithm called the Hitchcock bird-inspired algorithm (HBIA) is proposed. It is inspired by the aggressive bird behavior portrayed by Alfred Hitchcock in the 1963 thriller "The Birds." It is noteworthy to emphasize that the bird's behavior as shown in the movie is itself inspired by a considered natural birds behavior when faced with extreme conditions. HBIA is a stochastic swarm intelligence algorithm that captures the essence of the fictional behavior of the phenomenon of birds throughout the Hitchcock's film and model an optimization mechanism. The algorithm is based on the attack pattern of birds in the film, which has the stages of lurking, attack and reorganization, defined by the initialization, movement strategies in the search space and strategy of local minimum escape, respectively. The technique has as differential the use of adaptive parameters, a discretized random initialization and the use of the beta distribution. In contrast to the existing ones, the proposed technique provides an efficient optimization in high-dimensionality cost functions, using adaptive parameters, a discretized random initialization and the use of the beta distribution. Its performance is analyzed and compared to classic techniques, such as PSO, ABC and CS, as well as to the existing adaptive techniques, such as sine cosine algorithm, whale optimization algorithm, teaching–learning-based optimization and vortex search. HBIA's performance is investigated by several experiments implemented through eight cost functions. The results show that the HBIA can find more satisfactory solutions in large search spaces and high dimensionality of the evaluated cost functions when compared to the existing optimization methods. [ABSTRACT FROM AUTHOR]

In the presented paper, the functioning and the results of the work of two metaheuristic algorithms, namely cuckoo search algorithm (CS) and firefly algorithm (FA), are described using the apparatus of generalized nets (GNs), which is an appropriate and efficient tool for describing the essence of various optimization methods. The two developed GN-models mimic the optimization processes based on the nature of cuckoos and fireflies, respectively. The proposed GN-models execute the two considered metaheuristic algorithms conducting basic steps and performing optimal search. Building upon these two GN-models, a universal GN-model is constructed that can be used for describing and simulating both the CS and the FA by setting different characteristic functions of the GN-tokens. Moreover, the universal GN-model itself can be transformed to each of the herewith presented GN-models by applying appropriate hierarchical operators. In order to validate the proposed universal GN-model, numerical experiments are performed for the operating of the universal GN-model (CS and FA) on benchmark mathematical functions. The obtained results are compared with the results of the GN-model of CS, GN-model of FA, as well as the results of the standard CS and FA. [ABSTRACT FROM AUTHOR]

The simplicity and effectiveness of a recently proposed metaheuristic, butterfly optimization algorithm (BOA) have gained huge popularity among research community and are being used to solve optimization problems in various disciplines. However, the algorithm is suffering from poor exploitation ability and has a tendency to show premature convergence to local optima. On the other hand, the mutualism phase of another popular metaheuristic symbiosis organisms search (SOS) is known for its exploitation capability. In this paper, a novel hybrid algorithm, namely m-MBOA is proposed to enhance the exploitation ability of BOA with the help of mutualism phase of SOS. To evaluate the effectiveness of m-MBOA, thirty-seven (37) classical benchmark functions are considered and the performance of m-MBOA is compared with the performance of ten (10) state-of-the-art algorithms. Statistical tools have been employed to observe the efficiency of the m-MBOA qualitatively, and obtained results confirm the superiority of the proposed algorithm compared to the state-of-the-art metaheuristic algorithms. Finally, four real-life optimization problem, namely gear train design problem, gas compressor design problem, cantilever beam design problem and three-bar truss design problem are solved with the help of the newly proposed algorithm, and the results are compared with the obtained results of different popular state-of-the-art optimization techniques and found that the proposed algorithm is more efficient than the compared algorithms. [ABSTRACT FROM AUTHOR]

The identification of linear dynamic systems in the delta domain has been proposed in this paper with the help of a hybrid metaheuristic algorithm combining chaotic firefly algorithm (CFA) and grey wolf optimiser (GWO). GWO performs the global search, while CFA fine-tunes the solutions through its local search abilities, thereby balancing exploration and exploitation features. Linear systems with static nonlinearities at the input are termed as the Hammerstein model, whereas linear systems with static nonlinearities at the output are known as the Wiener model. A test case with continuous polynomial nonlinearities has been taken up for Hammerstein and Wiener system identification in the delta domain. Delta operator parameterisation unifies identification of continuous-time systems with the discrete domain at a higher sampling rate. Pseudo-random binary sequence (PRBS), polluted with white Gaussian noise of fixed signal-to-noise ratio (SNR), has been considered as the input signal to estimate the unknown model parameters as well as static nonlinear coefficients. The hybrid algorithm not only supersedes the parent heuristics of which it is constituted but also proves better in comparison with some standard and latest heuristic approaches reported in the literature. Nonparametric statistical tests are performed to validate the results. The plots of fitness function (normalised value) against the number of iterations also support the convergence speed and accuracy of the results. [ABSTRACT FROM AUTHOR]

In high-dimensional data, many of the features are either irrelevant to the machine learning task or are redundant. These situations lead to two problems, firstly overfitting and secondly high computational overhead. The paper proposes a feature selection method to identify the relevant subset of features for the machine-learning task using wrapper approach. The wrapper approach uses the Binary Bat algorithm to select the set of features and One-pass Generalized Classifier Neural Network (OGCNN) to evaluate the selected set of features using a novel fitness function. The proposed fitness function accounts for the entropy of sensitivity and specificity along with accuracy of classifier and fraction of selected features. The fitness function is compared using four classifiers (Radial Basis Function Neural Network, Probabilistic Neural Network, Extreme Learning Machine and OGCNN) on six publicly available datasets. One-pass classifiers are chosen as these are computationally faster. The results suggest that OGCNN along with the novel fitness function performs well in the majority of cases. [ABSTRACT FROM AUTHOR]

Given an edge-colored graph G, a tree with all its edges with different colors is called a rainbow tree. The rainbow spanning forest (RSF) problem consists of finding a spanning forest of G, with the minimum number of rainbow trees. In this paper, we present an integer linear programming model for the RSF problem that improves a previous formulation for this problem. A GRASP metaheuristic is also implemented for providing fast primal bounds for the exact method. Computational experiments carried out over a set of random instances show the effectiveness of the strategies adopted in this work, solving problems in graphs with up to 100 vertices. [ABSTRACT FROM AUTHOR]

Meta-heuristic algorithms are divided into two categories: biological and non-biological. Biological algorithms are divided into evolutionary and swarm-based intelligence, where the latter is divided into imitation based and sign based. The whale algorithm is a meta-heuristic biological swarm-based intelligence algorithm (based on imitation). This algorithm suffers from the early convergence problem which means the population convergences early to an unfavorable optimum point. Usually, the early convergence occurs because of the weakness in exploration capability (global search). In this study, an optimized version of the whale algorithm is proposed that introduces a new idea in grouping of whales (called GWOA) to overcome the early convergence problem. The proposed whale optimization algorithm is compared with the standard whale algorithm (WOA), CWOA improved whale algorithm, particle swarm optimization, and BAT algorithms applying CEC2017 functions. The results of the experiments show that the proposed method applying Friedman's test on 30 standard benchmark functions has a better performance than the other baseline algorithms. [ABSTRACT FROM AUTHOR]

Stochastic Fractal Search (SFS) is a novel and powerful metaheuristic algorithm. This paper presents a Multi-Objective Stochastic Fractal Search (MOSFS) for the first time, to solve complex multi-objective optimization problems. The presented algorithm uses an external archive to collect efficient Pareto optimal solutions during the optimization process. Using dominance rules, leader selection and grid mechanisms, MOSFS precisely approximates the true Pareto optimal front. The MOSFS is implemented on nine multi-objective benchmark functions (CEC 2009) with multimodal, convex, discrete and non-convex optimal Pareto fronts. Performance of the proposed algorithm is compared to well-known algorithms. In addition, different performance measures are considered to evaluate the convergence and coverage abilities of the algorithms including Inverted Generational Distance, Maximum Spread and Spacing. Furthermore, statistical analyses are utilized to determine the superior algorithm. The results revealed that the MOSFS performs significantly better than other algorithms in both convergence and coverage and it is able to approximate true Pareto front precisely. In the end, MOSFS is implemented to solve a real-world engineering design problem called welded beam design problem and efficiency of the algorithm is compared to recently developed algorithms. The results of simulations and the Wilcoxon rank-sum test showed that the MOSFS is able to provide the most promising Pareto front for the problem considering various performance metrics at a 95% confidence level. [ABSTRACT FROM AUTHOR]

Portfolio optimization is one of the most important problems in the finance field. The traditional Markowitz mean-variance model is often unrealistic since it relies on the perfect market information. In this work, we propose a two-stage stochastic portfolio optimization model with a comprehensive set of real-world trading constraints to address this issue. Our model incorporates the market uncertainty in terms of future asset price scenarios based on asset return distributions stemming from the real market data. Compared with existing models, our model is more reliable since it encompasses real-world trading constraints and it adopts CVaR as the risk measure. Furthermore, our model is more practical because it could help investors to design their future investment strategies based on their future asset price expectations. In order to solve the proposed stochastic model, we develop a hybrid combinatorial approach, which integrates a hybrid algorithm and a linear programming (LP) solver for the problem with a large number of scenarios. The comparison of the computational results obtained with three different metaheuristic algorithms and with our hybrid approach shows the effectiveness of the latter. The superiority of our model is mainly embedded in solution quality. The results demonstrate that our model is capable of solving complex portfolio optimization problems with tremendous scenarios while maintaining high solution quality in a reasonable amount of time and it has outstanding practical investment implications, such as effective portfolio constructions. [ABSTRACT FROM AUTHOR]

The detection of variables that contribute to the variation of a system is one of the most important considerations in the industrial manufacturing processes. This work presents the combination of Mahalanobis–Taguchi system and a hybrid binary metaheuristic based on particle swarm optimization and gravitational search algorithm (BPSOGSA) to perform an optimal feature selection in order to detect the relevant variables in a real process of foam injection in automotive industry. The proposed method is compared with other feature selection approach based in binary PSO algorithm. The experimental results revealed that BPSOGSA is faster and successfully converge selecting a smallest subset of features than BPSO. Moreover, the feature selection effect is validated through other widely used machine learning algorithms which improve their accuracy performance when they are trained with the subset of detected variables by the proposed system. [ABSTRACT FROM AUTHOR]

Nature-inspired heuristic and/or metaheuristic algorithms have been used for solving complex real-world problems in recent years. Electrohydrodynamic (EHD) inkjet printing is a microadditive manufacturing process in which high-resolution jets of polarizable functional materials were deposited on the defined spot of a substrate at the appointed time. The quality of the printed features is derived by the complex physics of the system. Parameter modeling of this process was carried out by using regression analysis, a feed-forward neural network trained with backpropagation (BPNN) and a neural network trained with a genetic algorithm (GA-NN) separately. This study emphasizes the droplet diameter prediction of an EHD inkjet printing system and explores the applicability of the soft computing-based methods for this new emerging technology. Soft computing-based approaches have been developed for the first time in this area to model the EHD inkjet process. Five hundred data were produced through the conventional regression analysis to train the neural network-based models. Output droplet diameter was predicted for different combinations of input parameters such as standoff height (SH), applied voltage (AV) and ink flow rate (FR) using the above three approaches, and their performances were analyzed through some randomly created real experimental test cases. All three models gave good prediction accuracy with less than 10% error in the prediction of the droplet diameter. Furthermore, it had been observed that the performance of GA-NN surpasses both the regression- and BPNN-based approaches in most of the test cases. It achieved quite satisfactory average absolute percentage deviation value of 2.51% between the target and predicted output using GA-NN model, which also showed an improvement over the regression or BPNN model. [ABSTRACT FROM AUTHOR]

Designing an ANN is a complex task as its performance is highly dependent on the network architecture as well as the training algorithm used to select proper synaptic weights and biases. Choosing an optimal design leads to greater accuracy when the ANN is used for classification. In this paper, we propose an approach multilayer perceptron-lion optimization algorithm (MLP-LOA) that uses lion optimization algorithm to find an optimum multilayer perceptron (MLP) architecture for a given classification problem. MLP-LOA uses back-propagation (BP) for training during the optimization process. MLP-LOA also optimizes learning rate and momentum as they have a significant role while training MLP using BP. LOA is a population-based metaheuristic algorithm inspired by the lifestyle of lions and their cooperative behavior. LOA, unlike other metaheuristics, uses different strategies to search for optimal solution, performs strong local search and helps to escape from worst solutions. A new fitness function is proposed to evaluate MLP based on its generalization ability as well as the network's complexity. This is done to avoid dense architectures as they increase chances of overfitting. The proposed approach is tested on different classification problems selected from University of California Irvine repository and compared with the existing state-of-the-art techniques in terms of accuracy achieved during testing phase. Experimental results show that MLP-LOA performs better as compared to the existing state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

This paper deals with hybrid flow shop scheduling problem with unrelated and eligible machines along with fuzzy processing times and fuzzy due dates. The objective is to minimize a linear combination of total completion time and maximum lateness of jobs. A mixed integer mathematical model is presented for the problem. The most challenging parts of hybrid evolutionary algorithms are determination of efficient strategies by which the whole search space is explored to perform local search around the promising search areas. In this study, a clustering-based approach as a data mining tool is introduced to identify the promising search areas. A repetitive clustering with an evolutionary algorithm is simultaneously employed to determine more promising parts of the solution space. Then, the searches in those parts are intensified with a local search. Here, two clustering-based meta-heuristic algorithms are applied to solve the problem, namely particle swarm optimization and genetic algorithm. The parameters are tuned by Taguchi experimental design, and various randomly generated test problems are used to evaluate the efficiency of the proposed algorithms. [ABSTRACT FROM AUTHOR]

Bat algorithm (BA) is a powerful nature-inspired swarm algorithm which finds applicability to a diverse range of problem domains. Though it is efficient, it suffers from two handicaps: possibility of being trapped in local optima and lost convergence speed as the algorithm progresses. This paper proposes swarm bat algorithm with improved search (SBAIS). SBAIS gains superior exploration capabilities by employing swarming characteristics inspired by shuffled complex evolution (SCE) algorithm. Best bats of the population are kept in a super-swarm, while all other bats are partitioned according to SCE. The super-swarm uses the search mechanism of bat algorithm with improved search to perform refined search around the best solution, which makes sure that the convergence speed of the algorithm is not lost. Every other swarm gets one solution from the super-swarm before starting their evolution process. These swarms evolve using standard bat algorithm, helping the algorithm to escape any possible local optima. SBAIS further keeps a check on the overall diversity of the population. If the diversity drops below a given threshold value, new random solutions are added to the population. Performance of SBAIS is validated by comparing it to BA and fourteen recent variants of bat algorithm over 30 standard benchmark optimization functions, CEC'05 and CEC'14 function sets. Results established the superiority of SBAIS over the compared algorithms. [ABSTRACT FROM AUTHOR]

Worker assignment is a relatively new problem in assembly lines that typically is encountered in situations in which the workforce is heterogeneous. The optimal assignment of a heterogeneous workforce is known as the assembly line worker assignment and balancing problem (ALWABP). This problem is different from the well-known simple assembly line balancing problem concerning the task execution times, and it varies according to the assigned worker. Minimal work has been reported in worker assignment in two-sided assembly lines. This research studies worker assignment and line balancing in two-sided assembly lines with an objective of minimizing the cycle time (TALWABP). A mixed-integer programming model is developed, and CPLEX solver is used to solve the small-size problems. An improved migrating birds optimization algorithm is employed to deal with the large-size problems due to the NP-hard nature of the problem. The proposed algorithm utilizes a restart mechanism to avoid being trapped in the local optima. The solutions obtained using the proposed algorithms are compared with well-known metaheuristic algorithms such as artificial bee colony and simulated annealing. Comparative study and statistical analysis indicate that the proposed algorithm can achieve the optimal solutions for small-size problems, and it shows superior performance over benchmark algorithms for large-size problems. [ABSTRACT FROM AUTHOR]

Up to now, many different methods (e.g., dispatching rules, metaheuristics) have been proposed to solve the job shop scheduling problem (JSSP), but the application of data mining approaches in the literature is limited. In this paper, we propose a data mining-based approach to generate an improved initial population for population-based heuristics/metaheuristics solving the JSSP. First, we apply a combination of 'attribute-oriented induction' and 'association rule mining' techniques to extract the rules behind the optimal or near-optimal schedules of JSSP. Then, a novel method called 'Assignment Procedure' is proposed to heuristically solve the JSSP using the extracted rules. The proposed method is able to generate numerous schedules for a given JSSP instance, and consequently, the generated solutions can be considered as the initial population for population-based solution methods. To evaluate the effectiveness of the proposed method, the two well-known metaheuristics are considered, i.e., genetic algorithm (GA) and particle swarm optimization (PSO). Extensive experiments were carried out on 35 benchmark problem instances, and Friedman test with post hoc Nemenyi test was performed to statistically analyze the results. The results revealed that using the initial populations generated by the proposed approach is very promising compared to the randomly generated population for both GA and PSO. Moreover, the experiments verify the significant amount of FEs that can be saved using the proposed approach and the superiority of the proposed method in comparison with the method of Koonce and Tsai (Comput Ind Eng 38:361–374, 2000) and four well-known dispatching rules in the literature. [ABSTRACT FROM AUTHOR]

In the paper, I Traffic flow guidance algorithm in intelligent transportation systems considering the effect of non i - I floating vehicle i , Chen et al. ([4]) present the method of estimating non-floating vehicles' driving information according to floating vehicles' information. The authors presented the estimation method, a new traffic flow guidance algorithm, Estimated Weighted Vehicle Density Feedback Strategy (EWVDFS) based on Weighted Vehicle Density Feedback Strategy (WVDFS). In the paper, A I robust lane detection method based on hyperbolic model i , Li et al. ([9]) investigated a robust lane detection method under structured roads. The paper I Road network i - I based region of interest mining and social relationship recommendation i by Tan and Zhang ([17]) proposed road context-based active region extraction algorithm (RAREA) which explores the method to extract the specific regions within the road network. [Extracted from the article]

As a critical research topic toward the new era of big data, how to develop a high-performance data analytics system has received significant research attention from different disciplines since the 2000s. In the literature, many recent works attempted to develop a high-performance data analytics system to handle the large amount of data (i.e., volume) from different information systems (i.e., variety) that typically will be created very quickly in a short time (i.e., velocity). In particular, several recent studies have shown that metaheuristic algorithms can be applied to many data mining optimization problems to provide a better way to find a high-quality result than traditional deterministic algorithms. A high-performance clustering algorithm for big data analytics system will be presented in this paper. The proposed algorithm is designed based on a new kind of metaheuristic algorithm, coral reef optimization with substrate layers (CRO-SL), to get a better cluster result. To improve the effectiveness and efficiency, the proposed CRO-SL scheme has been applied to a cloud computing platform as well to reduce the response time of a data analytics system. The simulation results show that the proposed algorithm is able to provide a better clustering result than the other clustering algorithms compared in this research, including k-means, genetic k-means algorithm, particle swarm optimization, and simple coral reef optimization algorithm in terms of the sum of squared errors. [ABSTRACT FROM AUTHOR]

In this paper, a new bat algorithm (BA) based on dynamic control parameters selection is presented. The dynamic BA (DBA) uses a new mechanism to dynamically select the best performing combination of the pulse rate coefficient, the pulse frequency coefficient, and the population size. A fractional-order PID (FOPID) controller based on the DBA is implemented to improve the performance of a distillation column process. The proposed FOPID controller is used to control the distillate and bottom mole fractions. The influence of the feed rate disturbance is considered for this model. The efficacy of the DBA-based FOPID is compared with the performance of the controllers based on the conventional BA, directional BA, enhanced BA, genetic algorithm, and particle swarm optimization algorithm. The analyses and simulation results show the superiority of the proposed method. [ABSTRACT FROM AUTHOR]

Plenty of problems are related to the calculation of edges and nodes in the realistic networks. It also influences the realization of shortest path problem (SPP) because of its essential fuzziness. This paper presents a fuzzy-based modified particle swarm optimization (fuzzy-based MPSO) algorithm for resolving the shortest path issue. The proposed work also evaluates the uncertainties of this shortest path problem through the utilization of offered algorithm. Actually, the normal PSO algorithm is altered and estimated to tackle the fuzzy-based SPP (FSPP) with uncertain edges. The performance of the planned algorithm will be improved; also the results are compared with the existing methodologies. The early convergence of the PSO technique can be alleviated and travelled via the dynamic operation of fuzzy method. And the proposed method is compared with other metaheuristic algorithms such as evolutionary random weight networks (GA-RWNs), grasshopper optimization algorithm with evolutionary population dynamics (GOA-EPD), levy weight grey wolf optimization (LGWO) and PSO in terms of cost and time consumption. The related results and discussion is performed in the working platform of MATLAB tool for the demonstration of the proposed work to manage the FSPP in indeterminate networks. [ABSTRACT FROM AUTHOR]

The consideration of this study is devoted to deal with the straight and U-shaped assembly line balancing problems (ALBPs). The ALBP involves allocation of required tasks to a set of workstations, so that objective functions being optimized are subjected to set of constraint. While many efforts have been dedicated in the literature to develop deterministic model of the assembly line, the attention is not considerably paid to those in uncertain circumstances. In this paper, along with proposing a novel fuzzy model for ALBP, triangular fuzzy numbers are deployed with to respect vagueness and uncertainty subjected to the task processing times. For this purpose, two conflicting objectives are considered simultaneously with regard to set of constraints, so that the efficiency of the line has to be maximized. To solve the problem, a modified NSGA-II, which utilized a new repairing mechanism, is proposed in response to the need of appropriate method treating such complicated problems. The validity of the proposed model and algorithm is evaluated and proved though a benchmark test problem. The obtained results reveal that in contrast to benchmark that applied an exact solution procedure, the proposed algorithm is capable of delivering the astonishing solutions in a more effective procedure. Along with the use of NSGA-II, in this study, three well-known meta-heuristic algorithms, namely PESA-II, NSACO and NPGA-II, are also employed for solving the problem in order to evaluate the effectiveness of the proposed algorithm, so that the results demonstrate the high performance for the NSGA-II over them. Finally, in light of the obtained results, this study offers an efficient framework enabling the decision maker to handle uncertainty in ALBPs along with the use of an efficient algorithm to solve them. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of fault diagnosis performed on a mechanical system, based on acquired vibrations from bearings. In this aim, an optimization algorithm resulted from the alliance between a time–frequency–scale signal processing method (the matching pursuit) and an evolutionary computing technique (mainly, a genetic algorithm) is introduced. The matching pursuit method itself leads to a NP-hard procedure, but, with the help of a metaheuristic, the procedure becomes computationally efficient. A generalization of Baker's procedure implementing the stochastic universal sampling mechanism, as well as a new concept, namely the Boltzmann annealing selection, is introduced, in order to design the genetic algorithm appropriately. This latter not only plays an important role in convergence speed, but also constitutes the basis of a (self) adaptive mechanism aiming to keep in balance the exploration and exploitation features. Based on the optimal solution found through the genetic matching pursuit procedure, the bearings fault diagnosis can successfully be performed, even in case of multiple defects and without prior training of some defect classification model. [ABSTRACT FROM AUTHOR]

Longest common subsequence (LCS) is a well-known NP-hard optimization problem that finds out the longest subsequence of each member of a given set of strings. In computational biology, sequence alignment is a fundamental technique to measure the similarity of biological sequences, such as DNA and genome sequences. A high sequence similarity often applied to molecular structural as well as functional similarities and can be used to determine whether (and how) sequences are related. Finding the longest common subsequence (LCS) is one way to measure the similarity of sequences. It has also applications in data compression, FPGA circuit minimization, and bioinformatics, etc. Exact algorithms are impractical since they fail to solve this problem for multiple instances of long lengths in polynomial time. There are some approximations, heuristic, and metaheuristic methods proposed to solve the problem. Chemical reaction optimization (CRO) is a new metaheuristic method that mimics the nature of chemical reaction into optimization problems. In this paper, we have proposed chemical reaction optimization technique to solve the longest common subsequence problem for multiple instances. Here, we have redesigned four elementary operators of CRO for LCS problem. Operators of CRO algorithm are used to explore the search space both locally and globally. A novel correction method has been designed to correct the solution. Correction method works after each search operator to ensure the validity of the changes made by operators. Both solution quality and execution time are considered while designing the operators and the correction method. Thus proposed system brings robustness, efficiency, and effectiveness while solving MLCS problem. Our approach is compared with hyper-heuristic, ant colony optimization, beam ant colony optimization, and memory-bound anytime algorithms. The experimental results in lengths of the returned common sequences show that our proposed algorithm gives either same or better results than all other algorithms in less execution time. [ABSTRACT FROM AUTHOR]

Differential evolution (DE) is one of the most powerful and effective evolutionary algorithms for solving global optimization problems. However, just like all other metaheuristics, DE also has some drawbacks, such as slow and/or premature convergence. This paper proposes a new subset-to-subset selection operator to improve the convergence performance of DE by randomly dividing target and trial populations into several subsets and employing the ranking-based selection operator among corresponding subsets. The proposed framework gives more survival opportunities to trial vectors with better objective function values. Experimental results show that the proposed method significantly improves the performance of the original DE algorithm and several state-of-the-art DE variants on a series of benchmark functions. [ABSTRACT FROM AUTHOR]

The forecasting of exchange rates has become a challenging area of research that has attracted many researchers over recent years. This work presents a sliding-window metaheuristic optimization-based forecast (SMOF) system for one-step ahead forecasting. The proposed system is a graphical user interface, which is developed in the MATLAB environment and functions as a stand-alone application. The system integrates the novel firefly algorithm (FA), metaheuristic (Meta) intelligence, and least squares support vector regression (LSSVR), namely MetaFA-LSSVR, with a sliding-window approach. The MetaFA automatically tunes the hyperparameters of the LSSVR to construct an optimal sliding-window LSSVR prediction model. The optimization effectiveness of the MetaFA is verified using ten benchmark functions. Two case studies on the daily Canadian dollar-USD exchange rate (CAN/USD) and the 4-h closing EUR-USD rates (EUR/USD) were used to confirm the performance of the system, in which the mean absolute percentage errors are 0.2532% and 0.169%, respectively. The forecast system has an 89.8–99.7% greater predictive accuracy than prior work when applied to the currency pair CAN/USD. With respect to the EUR/USD exchange rate, the error rates obtained using the proposed system were 20.8–23.9% better than those obtained by the baseline sliding-window LSSVR model. Therefore, the SMOF system is potentially useful for decision-makers in financial markets. [ABSTRACT FROM AUTHOR]

This paper explores rule decomposition for solving the MAXSAT problem. Four approaches are proposed to steer a bee swarm optimization metaheuristic. Two decomposition methods are proposed: direct and indirect. The first one applies the Kmeans algorithm, while the second one transforms a MAXSAT instance into a transactional database before performing decomposition using the Apriori algorithm. Several experiments conducted on DIMACS benchmark instances, and some other hard and large SAT instances have been carried out. Results show clear improvement compared to the state-of-the-art MAXSAT algorithms in terms of the quality of the obtained solutions. They show that the proposed approaches are stable when dealing with hard instances such as Parity8 from DIMACS. Results also demonstrate the superiority of the proposed approaches for medium and large instances. The proposed approaches could be applied to other optimization problems such as the weighted MAXSAT problem, the MAXCSP and coloring problems. They may also be adapted for other metaheuristics and decomposition methods. [ABSTRACT FROM AUTHOR]

Clustering is an unsupervised classification method used to group the objects of an unlabeled data set. The high dimensional data sets generally comprise of irrelevant and redundant features also along with the relevant features which deteriorate the clustering result. Therefore, feature selection is necessary to select a subset of relevant features as it improves discrimination ability of the original set of features which helps in improving the clustering result. Though many metaheuristics have been suggested to select subset of the relevant features in wrapper framework based on some criteria, most of them are marred by the three key issues. First, they require objects class information a priori which is unknown in unsupervised feature selection. Second, feature subset selection is devised on a single validity measure; hence, it produces a single best solution biased toward the cardinality of the feature subset. Third, they find difficulty in avoiding local optima owing to lack of balancing in exploration and exploitation in the feature search space. To deal with the first issue, we use unsupervised feature selection method where no class information is required. To address the second issue, we follow pareto-based approach to obtain diverse trade-off solutions by optimizing conceptually contradicting validity measures silhouette index (Sil) and feature cardinality (d). For the third issue, we introduce genetic crossover operator to improve diversity in a recent Newtonian law of gravity-based metaheuristic binary gravitational search algorithm (BGSA) in multi-objective optimization scenario; it is named as improved multi-objective BGSA for feature selection (IMBGSAFS). We use ten real-world data sets for comparison of the IMBGSAFS results with three multi-objective methods MBGSA, MOPSO, and NSGA-II in wrapper framework and the Pearson's linear correlation coefficient (FM-CC) as a multi-objective filter method. We employ four multi-objective quality measures convergence, diversity, coverage and ONVG. The obtained results show superiority of the IMBGSAFS over its competitors. An external clustering validity index F-measure also establish the above finding. As the decision maker picks only a single solution from the set of trade-off solutions, we employee the F-measure to select a final single solution from the external archive. The quality of final solution achieved by IMBGSAFS is superior over competitors in terms of clustering accuracy and/or smaller subset size. [ABSTRACT FROM AUTHOR]

In recent years, search-based software testing (SBST) is the active research topic in software testing. SBST is the process of generating test cases that use metaheuristics for optimization of a task in the framework of software testing to solve difficult NP-hard problems. The best fitness results must be found with the heuristic search among many possibilities for a more cost-effective testing process and automate the process of generating test cases. Although search-based test data generation is a field of interest, some challenges remain unknown. The main objective of this survey is to find the main topics and trends in this emerging field of search-based software testing by examining the methods and the literature of software testing. A review of earlier studies of search-based software testing from the year 1996 to 2016 is discussed with the application of metaheuristics for the optimization of software testing. [ABSTRACT FROM AUTHOR]

Driving trucks in a queue behind each other and in close proximity, called platooning, has been recently under consideration as a novel and promising approach to reduce fuel consumption, which provides environmental and financial benefits. This method works since driving in the slipstream of another vehicle reduces the aerodynamic drag, and as a result, less energy or fuel is consumed. This paper addresses this problem with the realistic assumptions of existing time constraints for trucks to depart from the origin and arrive at their destination, and waiting as well as detour possibility. As this problem is NP-hard even in its very simplified forms, a new meta-heuristic solution methodology inspired from ant colony optimisation is proposed to deal with it. Some sample problems of small to large size are generated and solved with our solution approach. The analysis of results shows the satisfactory performance of this meta-heuristic and its superiority over the exact and our previous approach with genetic algorithm. In addition, we analyse how the final result is affected by changing the main inputs and configurations of the problem. [ABSTRACT FROM AUTHOR]

This paper proposes a hybrid algorithm based on Water Cycle and Moth-Flame Optimization algorithms for solving numerical and constrained engineering optimization problems. The spiral movement of moths in Moth-Flame Optimization algorithm is introduced into the Water Cycle Algorithm to enhance its exploitation ability. In addition, to increase randomization in the new hybrid method, the streams in the Water Cycle Algorithm are allowed to update their position using a random walk (Levy flight). The random walk significantly improves the exploration ability of the Water Cycle Algorithm. The performance of the new hybrid Water Cycle-Moth-Flame Optimization algorithm (WCMFO) is investigated in 23 benchmark functions such as unimodal, multimodal and fixed-dimension multimodal benchmark functions. The results of the WCMFO are compared to the other state-of-the-art metaheuristic algorithms. The results show that the hybrid method is able to outperform the other state-of-the-art metaheuristic algorithms in majority of the benchmark functions. To evaluate the efficiency of the WCMFO in solving complex constrained engineering and real-life problems, three well-known structural engineering problems are solved using WCMFO and the results are compared with the ones of the other metaheuristics in the literature. The results of the simulations revealed that the WCMFO is able to provide very competitive and promising results comparing to the other hybrid and metaheuristic algorithms. [ABSTRACT FROM AUTHOR]

This paper presents a novel direct solution approach for fully fuzzy transportation problems in which all of the model parameters as well as decision variables are considered as fuzzy numbers. In detail, a fuzzy decoding procedure based on constrained fuzzy arithmetic operations and a fuzzy ranking technique is first introduced for solution of the problem directly without any fuzzy to crisp transformation process. Then, this decoding procedure is embedded into a metaheuristic, namely priority-based PSO algorithm for generating new solution vectors and seeking for better fuzzy acceptable solutions. By making use of the constrained fuzzy arithmetic concept, the proposed approach is also able to handle the decision maker's attitude toward risk. In order to show validity and applicability of the proposed approach, numerical examples on both balanced and unbalanced fully fuzzy transportation cases are generated and solved. The computational results have shown that relatively more precise and information efficient solutions can be obtained from the proposed approach for "risk-averse" and "partially risk-averse" decision makers. Furthermore, the proposed approach is also able to produce fuzzy solutions for "risk seekers" with high degree of uncertainty similar to the other methods available in the literature. [ABSTRACT FROM AUTHOR]

Differential evolution (DE) is a remarkable evolutionary algorithm for global optimization over continuous search space, whose performance is significantly influenced by its mutation operator and control parameters (scaling factor and crossover rate). In order to enhance the performance of DE, we adopt a novel Gaussian mutation operator and a modified common mutation operator to collaboratively produce new mutant vectors, and employ a periodic function and a Gaussian function to generate the required values of scaling factor and crossover rate, respectively. In the proposed variant of DE (denoted by GPDE), the two adopted mutation operators are adaptively applied to generate the corresponding mutant vector of each individual based on their own cumulative scores, the periodic scaling factor can provide a better balance between exploration ability and exploitation ability, and the Gaussian function-based crossover rate will possess fluctuant value, which possibly enhance the population diversity. To verify the performance of proposed GPDE, a suite of thirty benchmark functions and four real-world problems are applied to conduct the simulation experiment. The simulation results demonstrate that the proposed GPDE performs significantly better than five state-of-the-art DE variants and other two meta-heuristics algorithms. [ABSTRACT FROM AUTHOR]

Real-world problems are complex as they are multidimensional and multimodal in nature that encourages computer scientists to develop better and efficient problem-solving methods. Nature-inspired metaheuristics have shown better performances than that of traditional approaches. Till date, researchers have presented and experimented with various nature-inspired metaheuristic algorithms to handle various search problems. This paper introduces a new nature-inspired algorithm, namely butterfly optimization algorithm (BOA) that mimics food search and mating behavior of butterflies, to solve global optimization problems. The framework is mainly based on the foraging strategy of butterflies, which utilize their sense of smell to determine the location of nectar or mating partner. In this paper, the proposed algorithm is tested and validated on a set of 30 benchmark test functions and its performance is compared with other metaheuristic algorithms. BOA is also employed to solve three classical engineering problems (spring design, welded beam design, and gear train design). Results indicate that the proposed BOA is more efficient than other metaheuristic algorithms. [ABSTRACT FROM AUTHOR]

Population-based metaheuristic algorithms have been extensively applied to solve discrete optimization problems. Generally speaking, they work with a set of candidate solutions in the population which evolve during generations using variant reproduction and selection operations to find the optimal solution(s). The population is similar to a small society having several individuals which seek a common goal/solution. This study is motivated from the election systems of societies which can be applied in the population-based algorithms. We propose utilizing the majority voting for discrete population-based optimization algorithms which uses the information of all candidate solutions in the current generation to create a new trial candidate solution, called a president candidate solution. During optimization process, after applying the evolutionary operations, all candidate solutions vote collectively to determine the values of the president's variables. In the proposed method, a majority voting is utilized to choose a value for each variable (gene) of the president candidate solution. This method keeps untouched all other steps of population-based algorithms; therefore, it can be used with any kind of population-based algorithm. As case studies, the discrete differential evolution (DDE) algorithm and the discrete particle swarm optimization (DPSO) are used as the parent algorithms to develop majority voting-based discrete DE (MVDDE) and majority voting-based discrete PSO (MVDPSO). These two algorithms are evaluated on the fifteen discrete benchmark functions with the dimensions of D = 10, 30, 50, 100, 200 and 500. Simulation results confirm that majority voting-based discrete optimization algorithms obtain a promising performance on the majority of the benchmark functions. In addition, we have conducted some tests on large-scale 0-1 knapsack problems with large scales as a real-world application. [ABSTRACT FROM AUTHOR]

In this paper, we propose a multi-objective imperialistic competitive algorithm (MOICA) for solving global multi-objective optimization problems. The MOICA is a modified and improved multi-objective version of the single-objective imperialistic competitive algorithm previously proposed by Atashpaz-Gargari and Lucas (IEEE Congr Evolut Comput 7:4661-4666. doi:10.1109/CEC.2007.4425083, 2007). The presented algorithm utilizes the metaphor of imperialism to solve optimization problems. Accordingly, the individuals in a population are referred to as countries, of which there are two types—colonies and imperialists. The MOICA incorporates competition between empires and their colonies for the solution of multi-objective problems. To this end, it employs several non-dominated solution sets, whereby each set is referred to as a local non-dominated solution (LNDS) set. All imperialists in an empire are considered non-dominated solutions, whereas all colonies are considered dominated solutions. In addition to LNDS sets, there is one global non-dominated solution (GNDS) set, which is created from the LNDS sets of all empires. There are two primary operators in the proposed algorithm, i.e., assimilation and revolution, which use the GNDS and LNDS sets, respectively. The significance of this study lies in a notable feature of the proposed algorithm, which is that no special parameter is used for diversity preservation. This enables the algorithm to prevent extra computation to maintain the spread of solutions. Simulations and experimental results on multi-objective benchmark problems show that the MOICA is more efficient compared to a few existing major multi-objective optimization algorithms because it produces better results for several test problems. [ABSTRACT FROM AUTHOR]

Most of the meta-heuristic algorithms are based on the natural processes. They were inspired by physical, biological, social, chemical, social-biological, biological-geography, music, and hybrid processes. In this paper, artificial atom algorithm which was inspired by one of natural processes was applied to traveling salesman problem. The obtained results have shown that for small-scale TSP, artificial atom algorithm is closer to optimum than the other compared heuristic algorithms such as tabu search, genetic algorithm, particle swarm optimization, ant colony optimization, and their different combinations. [ABSTRACT FROM AUTHOR]

Behavioral teaching procedures can be used to promote the individualized learning of reading skills for children, and computational processes can assist instructors in the generation of a set of tasks. However, the automatic generation of these tasks can be unfeasible due to the high-order search space for the possible combinations of tasks; this complexity increases when considering the possible constraints as well as adapting the tasks to the individual characteristics of each student. This paper presents a new method to automatically generate teaching matching-to-sample tasks, adapting the difficulty by using bio-inspired optimization metaheuristics. Genetic algorithms, ant colony optimization, and integer and categorical particle swarm optimization were evaluated to determine their stability and capacity to generate adapted tasks. A comparison of the results between the algorithms showed a better rate of convergence for the genetic algorithms, which were able to generate tasks at an adapted level of difficulty to students. These tasks were applied to a group of students at a Brazilian public school in the early stages of a literacy course indicating satisfactory effects in the individual learning process. [ABSTRACT FROM AUTHOR]

Data clustering is one of the most popular branches of machine learning and data analysis. Partitioning-based type of clustering algorithms, such as K-means, is prone to the problem of producing a set of clusters that is far from perfect due to its probabilistic nature. The clustering process starts with some random partitions at the beginning, and then it attempts to improve the partitions progressively. Different initial partitions can result in different final clusters. Trying through all the possible candidate clusters for the perfect result is computationally expensive. Meta-heuristic algorithm aims to search for global optimum in high-dimensional problems. Meta-heuristic algorithm has been successfully implemented on data clustering problems seeking a near optimal solution in terms of quality of the resultant clusters. In this paper, a new meta-heuristic search method named elephant search algorithm (ESA) is proposed to integrate into K-means, forming a new data clustering algorithm, namely C-ESA. The advantage of C-ESA is its dual features of (i) evolutionary operations and (ii) balance of local intensification and global exploration. The results by C-ESA are compared with classical clustering algorithms including K-means, DBSCAN, and GMM-EM. C-ESA is shown to outperform the other algorithms in terms of clustering accuracy via a computer simulation. C-ESA is also implemented on time series clustering compared with classical algorithms K-means, Fuzzy C-means and classical meta-heuristic algorithm PSO. C-ESA outperforms the other algorithms in term of clustering accuracy. C-ESA is still comparable compared with state of art time series clustering algorithm K-shape. [ABSTRACT FROM AUTHOR]

In this paper, a novel meta-heuristic search algorithm inspired by rhinoceros’ natural behaviour is proposed, namely rhinoceros search algorithm (RSA). Similar to our earlier version called elephant search algorithm, RSA simplifies certain habitual characteristics of rhinoceros and stream-lines the search operations, thereby reducing the number of operational parameters required to configure the model. Via computer simulation, it is shown that RSA is able to outperform certain classical meta-heuristic algorithms. Different dimensions of optimization problems are tested, and good results are observed by RSA. The RSA is also implemented on permutation flow-shop scheduling problem (PFSP) with some representation method. Four different problem scales are used. Compared with partible swarm optimization (PSO) on PFSP, the RSA outperforms PSO on different problem scales with a 3% improvement. [ABSTRACT FROM AUTHOR]

Deep learning-based approaches have been paramount in recent years, mainly due to their outstanding results in several application domains, ranging from face and object recognition to handwritten digit identification. Convolutional neural networks (CNNs) have attracted a considerable attention since they model the intrinsic and complex brain working mechanisms. However, one main shortcoming of such models concerns their overfitting problem, which prevents the network from predicting unseen data effectively. In this paper, we address this problem by means of properly selecting a regularization parameter known as dropout in the context of CNNs using meta-heuristic-driven techniques. As far as we know, this is the first attempt to tackle this issue using this methodology. Additionally, we also take into account a default dropout parameter and a dropout-less CNN for comparison purposes. The results revealed that optimizing dropout-based CNNs is worthwhile, mainly due to the easiness in finding suitable dropout probability values, without needing to set new parameters empirically. [ABSTRACT FROM AUTHOR]