Search

Most existing evolutionary algorithms encounter difficulties in dealing with multi-objective problems (MOPs) whose Pareto optimal solutions are sparse, especially when the decision space of the MOP is large. Most existing sparse evolutionary algorithms (EAs) lack the tricks specially for large-scale problems, and the existing tricks need some grouping/variable analysis methods to divide the large-scale variables, which severely affects the effect of optimization. To fill this gap, an evolutionary algorithm based on rank-1 approximation, suitable to reduce the dimensionality of sparse higher-order datasets, is proposed. First, canonical polyadic decomposition is introduced to divide the higher-dimensional set of decision variables into several lower-dimensional sub-sets. These sub-populations are alternatively optimized by a sparse evolution algorithm with special genetic strategies, to search for improved solutions in their respective lower-dimensional decision sub-space. Finally, the whole population is asymptotically optimized by all the sub-populations. The proposed algorithm first introduces the rank-1 approximation into the dimensionality reduction in sparse large-scale variables, combines it with the sparse EA, and proposes efficient alternative optimization strategy to solve the non-convex problem resulted from the multiple sub-problems. It is compared with other state-of-the-art algorithms for large-scale and sparse problems. The experimental results indicate that it outperforms other algorithms in terms of optimization performance and convergence rate. [ABSTRACT FROM AUTHOR]

Meta-heuristics have been successfully applied to many complex optimization problems. One of the main reasons for its success is its ability to handle non-convex, nonlinear, multimodal, multi-variable, and multi-objective problems with easy implementation. However, the quality of the response of these algorithms to an optimization problem is highly susceptible to the control parameters, and few works aim to tune them or find tools that can improve the algorithms. The literature is rich in proposals for new algorithms, but not for improving existing ones. This paper presents different strategies for tuning and accelerating meta-heuristics using the first hybrid algorithm in the literature. The Lichtenberg algorithm is inspired by lightning and Lichtenberg figures and has been increasingly successfully applied to various optimization problems. However, a study of its best parameters has never been presented until now. After a discussion of the best tuning tools, its tuning parameters are performed using response surface methodology. Then, 14 versions are studied through 10 test functions using chaos theory and Lévy flights scenarios. After 13,500 simulations, the chaotic Lichtenberg algorithm equipped with the piecewise function and tuned parameters proved the best version with only 16% similarity to the original algorithm. Then, it was compared to the genetic algorithm, particle swarm optimization, gray wolf optimizer, salp swarm optimization, whale optimization algorithm, and dragonfly algorithm. The proposed algorithm had both the best average accuracy, lower computational cost, and the smallest standard deviation. [ABSTRACT FROM AUTHOR]

Empirical distribution-based framework (EDBF) is employed to lots of multi-parent crossover algorithms (MCAs) to make it much more efficient to spawn coefficients within a certain boundary at each iteration. However, EDBF cannot work with the numerous parent chromosomes especially exceeding fifty. To address this problem, an enhanced EDBF framework, namely adaptive boundary constraint framework (ABCF), is proposed by adaptively changing the boundary of weight assigned to each parent chromosome according to the constraint law. Furthermore, ABCF is compared with EDBF, ABPSO and RCBBFA algorithms on 20 benchmark functions in terms of convergence, efficiency and accuracy. Experimental results demonstrate that ABCF outperforms comparative algorithms on most of benchmark functions. As a general framework rather than a specific algorithm, ABCF is easy to implement and can easily be accommodated to any existing MCAs. [ABSTRACT FROM AUTHOR]

A Hybrid Quantum Genetic Algorithm with an Adaptive Rotation Angle (HQGAAA) for the 0-1 knapsack problem is presented. This novel proposal uses the Deutsch-Jozsa quantum circuit to generate quantum populations, which synergistically works as haploid recombination and mutation operators taking advantage of quantum entanglement providing exploitative and explorative features to produce new individuals. Furthermore, the created individuals are updated using an adaptive rotation angle operator that helps refine new individuals to converge to the optimal solution. We performed comparative tests with other quantum evolutionary algorithms and the classical genetic algorithm to demonstrate that this proposal performed better with the tested problem. Results showed that quantum algorithms performed similar but better than the classic genetic algorithm regarding accuracy. Moreover, statistic tests demonstrated that our proposal is faster than the other quantum algorithms tested. [ABSTRACT FROM AUTHOR]

Most genetic operators use random mating selection strategy and fixed rate crossover operator to solve various optimization problems. In order to improve the convergence and diversity of the algorithm, an adaptive adjacent maximum distance crossover operator is proposed in this paper. A new mating selection strategy (distance-based mating selection strategy) and an adaptive mechanism (adaptive crossover strategy based on population convergence) are adopted. Distance-based mating selection strategy purposefully selects parents to produce better offspring. Adaptive crossover strategy based on population convergence increases the convergence speed of the algorithm by controlling the crossover probability. The proposed crossover strategy is evaluated on the simulated binary crossover operators of non-dominated sorting genetic algorithm II and multi-objective evolutionary algorithm based on decomposition. The performance of the algorithm is verified on a series of standard test problems. Finally, the optimization results of the improved algorithm using adaptive adjacent maximum distance crossover operator and the standard algorithm are compared and analyzed. The experimental results show that the algorithm using adaptive adjacent maximum distance crossover operator has better optimization results. [ABSTRACT FROM AUTHOR]

The purpose of this article is to develop a competitive supply chain network (SCN) in the face of uncertainty. The objective of the leader chain is to maximize total network profits by strategically locating suppliers, manufacturers, distribution centers, and retailers. Additionally, the follower chain seeks to maximize the network's profit. Both factors, optimal flow allocation to different echelons of the SCN and product pricing, are examined in the leader chain and follower chain. The KKT conditions are used in this article to convert a bi-level model to a one-level model. Additionally, a fuzzy programming technique is used to control the problem's uncertain parameters. According to the results obtained using the fuzzy programming technique, increasing the uncertainty rate increases demand while decreasing the OBFV and average selling price of products. Finally, the problem was untangled using a novel hybrid genetic and ant-lion optimization algorithm (HGALO). The results of problem solving in larger sizes demonstrate HGALO's superior efficiency in comparison with the other algorithm. [ABSTRACT FROM AUTHOR]

Many new, nature-inspired optimization algorithms are proposed these days, and these algorithms are gaining popularity day by day. These algorithms are frequently preferred for these real-world problems as they need less information, are reliable and robust, and have a structure that can easily be applied to discrete problems. Too many algorithms result in difficulty choosing the correct technique for the problem, and selecting an unwise method affects the solution quality. In addition, some algorithms cannot be reliable for some specific real-world problems but very successful for others. In order to guide and give insight into the practitioners and researchers about this problem, studies involving the comparison and evaluation of the performance of algorithms are needed. In this study, the performances of six nature-inspired methods, which included five new implementations of differential evolutionary algorithms (DE), scatter search (SS), equilibrium optimizer (EO), marine predators algorithm (MPA), and honey badger algorithm (HBA) applied to land redistribution problem and genetic algorithms (GA), were compared. In order to compare the algorithms in detail, various performance indicators were used as problem based and algorithm based. Experimental results showed that DE and SS algorithms have a more successful performance than the other methods by solution quality, robustness, and many problem-based indicators. [ABSTRACT FROM AUTHOR]

The objective of a fixed-cost transportation problem (FCTP) is to minimize the total transportation cost (or to maximize the total profit). FCTPs become more complex due to the presence of uncertain factors like suppliers' capacity, customers' demand, fixed cost and amount of goods transported. In order to deal with its complexity, three mathematical models, namely the expected value, chance-constrained and measure chance models, are basically available in the literature. This paper aims to solve the crisp equivalent models of an uncertain two-echelon FCTP by employing a robust hybrid evolutionary algorithm. The framework of the designed algorithm comprises of parameter-free Jaya algorithm and the efficient quadratic approximation operator. Initially, its efficiency is validated on the 24 constrained benchmark functions from CEC 2006 (Liang et al. in Problem definitions and evaluation criteria for the CEC 2006 special session on constrained real-parameter optimization, 2006) and 30 unconstrained benchmark functions (Rao in Int J Ind Eng Comput 7:19–34, 2016). Friedman rank test and Iman–Davenport test are conducted for both constrained and unconstrained test instances. Additionally, five real-life engineering design problems are solved to further analyze the flexibility of the proposed algorithm. The computational results and convergence graphs justify the robustness of the designed algorithm over state-of-the-art algorithms. Finally, its application is demonstrated on chance-constrained and measure chance two-echelon FCTP models. Near-optimal results are contradicted by genetic algorithm and particle swarm optimization in solving both these models. Further, a few large-scale instances of FCTP are solved by the proposed algorithm. [ABSTRACT FROM AUTHOR]

Grasshopper optimization algorithm (GOA) is one of the promising optimization algorithms for optimization problems. However, it has the main drawback of trapping into a local minimum, which causes slow convergence or inability to detect a solution. Several modifications and combinations were suggested to overcome this problem. This paper presents a modified grasshopper optimization algorithm (MGOA)-based genetic algorithm to overcome this problem. Modifications rely on certain mathematical assumptions and varying the domain of the control parameter, Cmax, to escape from the local minimum and move the search process to an improved point. Parameter C is one of the essential parameters in GOA, where it balances the exploration and exploitation of the search space. These modifications aim to speed up the convergence rate by reducing the repeated solutions and the number of iterations. Both the original GOA and the proposed algorithms are tested with 19 main test functions to investigate the influence of the proposed modifications. In addition, the algorithm will be applied to solve five different cases of nonlinear systems with different types of dimensions and regularity to show the reliability and efficiency of the proposed algorithm. Promising results are achieved compared to the original GOA. The proposed approach shows an average percentage of improvement of 96.18 as illustrated in the detailed results. [ABSTRACT FROM AUTHOR]

With the growth of multinational companies, increasing international and domestic competition between companies, upgrading information technology, and increasing customer expectations, accurate supply chain (SC) planning is essential. In such an environment, pollution has become more severe in recent decades, and with the weakening of the environment and global warming, green SC management strategies have become significant issues in recent decades. In this research, we consider the integrated production and distribution planning problem of a multi-level green closed-loop SC system, which includes multiple recycling, manufacturing/ remanufacturing, and distribution centers. We present a three-level bi-objective programming model to maximize profit and minimize the amount of greenhouse gas emissions. A hierarchical iterative approach utilizing the LP-metric method and the non-dominated sorting genetic algorithm (NSGA-II) is introduced to solve the proposed model. The Taguchi approach is applied to find optimum control parameters of NSGA-II. Moreover, Monte Carlo (MC) simulation is applied to tackle uncertainty in demand, and the NSGA-II algorithm is integrated with MC simulation (MCNSGA-II). Through numerical experiments on randomly generated instances, we observe that the average of the relative gaps NSGA-II and LP-metric method are 7%, 7%, and 1% for the three levels, respectively. Furthermore, it is observed that the value of convergence (C) and spacing (S) metrics of the MCNSGA-II algorithm are 0 and 1.82E + 07, which are better than NSGA-II. [ABSTRACT FROM AUTHOR]

Heart disease is the most life-threatening disease globally, affecting human life very critically. On-time and precise diagnosis of heart disease is vital for the prevention and treatment of heart failure. In the earlier stage, several detection techniques were developed to detect heart disease using machine learning techniques. But the heart problems have not been detected accurately with the minimum required time. This paper presented an effective heart disease prediction system that accurately diagnoses abnormality within a short duration. The feature selection process is carried with stochastic gradient boosting with a recursive feature elimination approach. For classification, the features are clustered by using the adaptive Harris hawk optimization clustering approach. Based on clustered features, the classification is processed with an enhanced deep genetic algorithm. This method improves the performance of the deep neural network by augmenting its initial weights with an enhanced genetic algorithm, which recommends better weights for the neural network. The demonstration of the proposed work is simulated in the MATLAB platform and the dataset used in UCI machine learning repository. The experimental outcome shows that the developed feature selection and classification approach attain 98.36% and 97.3% accuracy compared with the state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

In this paper, a new U-shaped assembly line balancing problem is studied. For the first time, the criteria such as equipment cost, number of stations and activity performing quality level are considered to be optimized simultaneously by activity to station and worker to station decisions. For this aim, a multi-objective nonlinear formulation is proposed and its linearized version is also presented. Since, according to the literature, the U-shaped assembly line balancing problem with equipment requirements is an NP-hard problem, the problem of this study is NP-hard too. Because of this complexity, the classical algorithms like simulated annealing, variable neighborhood search, and classical genetic algorithm with a novel encoding/decoding scheme are used as solution approaches. As an extension, two hybrid versions of the proposed classical algorithms are proposed according to the characteristics of the problem. In order to evaluate the proposed meta-heuristics, because the problem is new, some test problems are generated randomly. Computational study of the paper, including sensitivity analysis of the proposed meta-heuristics and final experiments on the test problems, proves the superiority of the hybrid versions of the classical algorithms. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel hybrid metaheuristic for the Knapsack Problem with Forfeits (KPF). KPF is a recently introduced generalization of the Knapsack Problem. In this variant, a penalty cost incurs whenever both items composing a so-called forfeit pair belong to the solution. Our proposed algorithm, called GA–CG Forfeits, combines the strengths of the Genetic and Carousel Greedy paradigms. In this work, we define the algorithm and compare it with two previously proposed heuristics on a set of benchmark instances. In these tests, GA–CG Forfeits provided significantly better solutions than the other two algorithms on all instances. [ABSTRACT FROM AUTHOR]

A large amount of data available on Web has proven to be an immense resource for innovative recommender system (RS) techniques and concepts. The traditional recommender system intended to provide recommendations for a single user. However, in certain domains the recommendation is required for a group of users. As to provide better recommendations for a group of users, we leverage the concept of circles in a network. In this work, we use the genetic algorithm (GA) K_Means clustering algorithm to generate social circles in a network. Then, we compute the status of each user in these overlapping circles. Finally, a circle-based group recommendation approach is used to generate the final group recommendation. The results obtained on the Epinions dataset validate the eminence of the proposed model over traditional approaches of group recommendation. [ABSTRACT FROM AUTHOR]

One of the most effective corrective control strategies to prevent voltage collapse and instability is load shedding. In this paper, a multiple-deme parallel genetic algorithm is used for a suitable design of load shedding. The load shedding algorithm is implemented when the voltage stability margin index of the power system is lower than a predefined value. In order to increase the computational speed, the voltage stability margin index is estimated by a modular neural network method in a fraction of a second. In addition, in order to use the exact values of the voltage stability margin index for neural network training, a simultaneous equilibrium tracing technique has been employed considering the detailed model of the components of the generating units such as the governor and the excitation system. In the proposed algorithm, the entire population is partitioned into several isolated subpopulations (demes) in which demes distributed in different processors and individuals may migrate occasionally from one subpopulation to another. The proposed technique has been tested on New England-39 bus test system, and the obtained results indicate the efficiency of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper, the load forecasting model is established to increase the precision of meteorological impacts, temperature, short-term power load forecasting, working and holiday factors by considering the power load. Further, IGA-LS-SVM is proposed which is a short-term power load forecasting technique based on AI algorithm. And to increase the forecast accuracy and generalization capability of LS-SVM, we applied the adopted mutation probability and new coding technology to the parameter optimization of LS-SVM. The temperature, load, weather state, working and holiday days be taken as prediction model as input, and load value was predicted output. We selected the sample data from meteorological information and historical load of a city in Yunnan province. By results, the prediction verifies the good prediction effect when associated with existing BP algorithm and the proposed IGA- LS-SVM algorithm yields a value 0.8274 more significant than all others, which is appropriate for short-term power load prediction. [ABSTRACT FROM AUTHOR]

In construction industry, general contractors (GCs) need to manage and conduct numerous projects simultaneously. To this aim, they usually have subcontractors conduct available tasks and projects. So, subcontractor management is becoming a major challenge. To deal with this challenge, GCs aim to reduce total costs of projects including employment/subcontracting costs, indirect costs, tardiness penalties, and the money which must be paid for movement of workforce from one project to another. The aim of current research study is to propose a model for selecting subcontractors and assigning available tasks in the project to them in order to reduce the costs of the GC. Then, a genetic algorithm is proposed to solve a real problem. The proposed algorithm is innovative from three points of view: (1) generation of initial population; (2) subcontractor assignment approach; and (3) the objective function. The problem was also solved by means of an exact method. Then, the results of proposed algorithm were compared to the outcomes of the exact method. This comparison shows that the proposed algorithm can efficiently help GCs select subcontractors and assign available tasks to each subcontractor when several projects must be carried out simultaneously. [ABSTRACT FROM AUTHOR]

Today, there are a lot of useful algorithms for covering array (CA) generation, one of the branches of combinatorial testing. The major CA challenge is the generation of an array with the minimum number of test cases (efficiency) in an appropriate run-time (performance), for large systems. CA generation strategies are classified into several categories: computational and meta-heuristic, to name the most important ones. Generally, computational strategies have high performance and yield poor results in terms of efficiency, in contrast, meta-heuristic strategies have good efficiency and lower performance. Among the strategies available, some are efficient strategies but suffer from low performance; conversely, some others have good performance, but is not such efficient. In general, there is not a strategy that enjoys both above-mentioned metrics. In this paper, it is tried to combine the genetic algorithm and the Augmented Lagrangian Particle Swarm Optimization with Fractional Order Velocity to produce the appropriate test suite in terms of efficiency and performance. Also, a simple and effective minimizing function is employed to increase efficiency. The evaluation results show that the proposed strategy outperforms the existing approaches in terms of both efficiency and performance. [ABSTRACT FROM AUTHOR]

The swallow swarm optimization (SS) is a challenging method of optimization, which has a quicker convergence speed, not getting caught in the local extreme points. However, the SS suffers from a few shortcomings—(1) the movement speed of particles is not controlled suitably during the search due to the requirement of an inertia weight and (2) the less flexibility of variables does not permit to maintain a balance between the local and the global searches. To solve these problems, a new Levy swallow swarm optimization (SSLY) algorithm with the exploitation capability is proposed. This article also provides an optimal design methodology for the low-pass filter using the suggested SSLY technique. A new objective function is introduced to achieve the maximally flat frequency response, which is another important contribution to the field. The firefly algorithm (FA), the sine cosine algorithm (SCA) and the standard global optimizers—real coded genetic algorithm (GA), conventional particle swarm optimization (PSO), cuckoo search (CS) and SS, are considered for a comparison. The proposed SSLY outperforms the FA, SCA, GA, PSO, CS and SS algorithms. Results authenticate suitability of the proposed algorithm for solving the filter design problems in the FIR domain. [ABSTRACT FROM AUTHOR]

Cloud computing delivers practical solutions for long-term image archiving systems. Cloud data centers consume enormous amounts of electrical energy that increases their operational costs. This shows the importance of investing on energy consumption techniques. Dynamic placement of virtual machines to appropriate physical nodes using metaheuristic algorithms is among the methods of reducing energy consumption. In metaheuristic algorithms, there should be a balance between both exploration and exploitation aspects so that they can find better solutions in a search space. Exploration means looking for a solution in a wider area, while exploitation is producing new solutions from existence ones. Artificial bee colony optimization, which is a biological metaheuristic algorithm, is a sign-oriented approach. It has a strong exploration ability, but a relatively weaker exploitation power. On the other hand, tabu search is a popular algorithm that shows better exploitation in comparison with ABC. In this study, cloud computing environments are detailed with an allocation protocol for efficient energy and resource management. The technique of energy-aware allocation splits data centers (DCs) resources among client applications end routes to enhance energy efficacy of DCs and also achieves anticipated quality of service (QoS) for everyone. Heuristic protocols are exercised for optimizing the distribution of resources to upgrade the efficiency of DC. In the current paper, energy-aware resources allotment technique is employed and optimized in clouds via a new approach called Tabu Job Master (JM). Tabu JM claims the benefits of some variables and also rapid convergence speeds. Results are duly achieved for energy consumption—the count of virtual machines (VMs) migration and also makespan. The results shown by Tabu JM are benchmarked by using genetic algorithm (GA), artificial bee colony (ABC), ABC with crossover and technique of mutation, the basic tabu search techniques, and Tabu Job Master. [ABSTRACT FROM AUTHOR]

Forming a team of experts that can match the requirements of a collaborative task is an important aspect, especially in project development. In this paper, we propose an improved Jaya optimization algorithm for minimizing the communication cost among team experts to solve team formation problem. The proposed algorithm is called an improved Jaya algorithm with a modified swap operator (IJMSO). We invoke a single-point crossover in the Jaya algorithm to accelerate the search, and we apply a new swap operator within Jaya algorithm to verify the consistency of the capabilities and the required skills to carry out the task. We investigate the IJMSO algorithm by implementing it on two real-life datasets (i.e., digital bibliographic library project and StackExchange) to evaluate the accuracy and efficiency of proposed algorithm against other meta-heuristic algorithms such as genetic algorithm, particle swarm optimization, African buffalo optimization algorithm and standard Jaya algorithm. Experimental results suggest that the proposed algorithm achieves significant improvement in finding effective teams with minimum communication costs among team members for achieving the goal. [ABSTRACT FROM AUTHOR]

In the ensemble feature selection method, if the weight adjustment is performed on each feature subset used, the ensemble effect can be significantly different; therefore, how to find the optimized weight vector is a key and challenging problem. Aiming at this optimization problem, this paper proposes an ensemble feature selection approach based on genetic algorithm (EFS-BGA). After each base feature selector generates a feature subset, the EFS-BGA method obtains the optimized weight of each feature subset through genetic algorithm, which is different from traditional genetic algorithm directly processing single features. We divide the EFS-BGA algorithm into two types. The first is a complete ensemble feature selection method; based on the first, we further propose the selective EFS-BGA model. After that, through mathematical analysis, we theoretically explain why weight adjustment is an optimization problem and how to optimize. Finally, through the comparative experiments on multiple data sets, the advantages of the EFS-BGA algorithm in this paper over the previous ensemble feature selection algorithms are explained in practice. [ABSTRACT FROM AUTHOR]

Precise characterization of underground reservoirs requires accurate calculations of the reservoir's petrophysical data and accurate selection of the mathematical model governing the reservoir's dynamic. In this study, we develop a novel heuristic-clustering algorithm, namely GA–DBSCAN–KMEANS, that can be applied over pressure transient data to assess the true reservoir model out of a pool of candidates. In this algorithm, each specific reservoir model is considered a subpopulation in the GA (genetic algorithm). Then, the simultaneous optimization of all the reservoir models is sought using the proposed hybrid algorithm. During the optimization process, the population size of different models will be either decreased, increased, or unchanged based on the average quality match obtained for each model. A combined DBSCAN (density-based spatial clustering of applications with noise)–KMEANS clustering scheme is used to increase the population size for the best reservoir model in each iteration of the GA. The accuracy of the proposed algorithm was verified using several synthetic data and a real field case obtained from the open literature. The tested data were collected from different types of reservoir models, including homogeneous reservoirs, matrix-fracture dual-porosity reservoirs, and fault-limited reservoirs. For uncertainty analyses and to test the performance of the algorithm under large numbers of initializations, Monte Carlo simulations were conducted. Results of the Monte Carlo simulations unveiled high values of P10, P50, and P90 for the probability of the true reservoir model and low values of these statistics for the false reservoir models. This shows that the outcome of the proposed algorithm is not affected by the initial randomization of the solution subspaces; hence, the developed algorithm is a reliable tool in determining the most probable reservoir model from transient well testing data. [ABSTRACT FROM AUTHOR]

The research paper presents implementation of a fuzzy rule and membership function-based fuzzy-aided PID controller for automatic generation control (AGC) in multiarea nonlinear power system. At the initial stage of this proposed work, a three-area nine-unit installed interconnected network is considered for developing different dynamic responses in response to AGC analysis. A modified approach named modified sine cosine algorithm (M-SCA) is proposed for tuning the gain parameters of the above-proposed fuzzy controller to produce close optimum gain values. The proposed modified algorithm is developed from its original sine cosine algorithm by improving and updating few equations which is capable of making the balance between exploration and exploitation levels of this algorithm and improving the updating quality of iteration. To impose supremacy of M-SCA technique, it is examined through convergence curves and its performance is compared with host sine cosine algorithm, genetic algorithm, and particle swarm optimization algorithm. For controller supremacy analysis, the performance of the proposed fuzzy-aided PID controller is compared with conventional I, PI, and PID controllers, and it has been revealed that proposed M-SCA-tuned fuzzy-aided PID controller exhibits better performances through different deviated responses for AGC analysis. To demonstrate most standard and supremacy of proposed approaches, finally these are tested through a five-area ten-unit system considering some physical nonlinear constraints like generation rate constraint, governor dead band, boiler dynamics and time delay. At the final observation level, the proposed fuzzy controller has gone through different sensitivity analyses with variation of different system parametric conditions and different load conditions. [ABSTRACT FROM AUTHOR]

Feature selection is one of the hottest machine learning topics in recent years. The main purposes of it are to simplify the original model, improve the readability of the model, and prevent over-fitting by searching for a suitable subset of features. There are many methods for this problem, including evolutionary algorithms and particle swarm optimization. Among them, the competitive swarm optimizer is a new optimization algorithm proposed in recent years, which is based on particle swarm optimization algorithm, and has achieved good results in high-dimensional feature selection problems, but it also has the problems of high computation time cost and easily being premature. Aiming at these problems, this paper proposes to add the crossover operator and mutation operator in the genetic algorithm to the competitive swarm optimization, so as to improve the generation speed of new individuals in the algorithm and prevent premature population. After testing on UC Irvine Machine Learning Repository, the new algorithm not only improves the computational efficiency, but also avoids the problem that the competitive swarm optimization algorithm is easy to fall into the local optimum, which greatly improves the calculation effect. [ABSTRACT FROM AUTHOR]

With the advent of neural networks, fuzzy logic, genetic algorithms, and other soft computing methodologies, many researchers have demonstrated successful designs for intelligent control strategies that appear to outperform traditional linear feedback controls; however, industry has mostly ignored the new technology due to the lack of stability guarantees (required in most formal engineering risk-management processes). In this paper, we offer a Lyapunov-stability framework where one can place any arbitrary computational algorithm and still get guarantees of uniformly ultimately bounded (UUB) signals. We would expect an intelligent algorithm to be well-designed such that the proposed framework would not come into play unless unanticipated disturbances affect the system. But even if the intelligent algorithm was poorly designed, the resulting performance (inside our framework) would just look similar to that of a typical nonlinear neural-adaptive control. In our strategy, the intelligent algorithm trains a cerebellar model articulation controller (CMAC) neural network with arbitrarily bounded weights in order to achieve good performance, while a second CMAC trains in parallel using direct-adaptive-control laws in order to provide stability (even in the case of the first CMAC weights reaching their imposed bound). We test the strategy using a previously proposed ad hoc CMAC weight smoothing strategy, serving as the intelligent algorithm, with simulations and experiment controlling a two-link flexible-joint robot. [ABSTRACT FROM AUTHOR]

This paper presents a new variant of particle swarm optimization (PSO) algorithm named guided adaptive search-based particle swarm optimizer (GuASPSO). In this algorithm, the personal best particles are all divided into a linearly decreasing number of clusters. Then, the unique global best guide of a given particle located at a cluster is obtained as the weighted average calculated over other clusters' best particles. Since the clustered particles are being well-distributed over the whole search space in the clustering process, there would be a moderate distance between each particle and its unique global best guide, contributing the particles neither to be trapped in local optima nor engaged in a drift leading to lose diversity in the search space. In this approach, the number of clusters is high at the early iterations and is gradually decreased by lapse of iterations to less stress the diversity factor and further stress the fitness role to cause the particles to better converge to the optimal point. Holding this balance between global and personal bests' role to attract the particles, on the one hand and between convergence and diversity, on the other hand, can hold a better exploration–exploitation balance in the proposed algorithm. To test the performance of GuASPSO, four popular meta-heuristic algorithms, including genetic algorithm, gravitational search algorithm, gray wolf optimizer, and PSO algorithm as well as 23 standard benchmark functions as the test beds, are employed. The experimental results validated GuASPSO as a robust well-designed algorithm to handle various optimization problems. [ABSTRACT FROM AUTHOR]

The Steiner minimal tree (SMT) problem is an NP-hard problem, which is the best connection model for a multi-terminal net in global routing problem. This paper presents a unified algorithm for octagonal and rectilinear SMT construction based on hybrid transformation strategy (HTS) and self-adapting particle swarm optimization. Firstly, an effective HTS is proposed to enlarge the search space and improve the convergence speed. Secondly, the proposed HTS in the evolutionary process may produce an ineffective solution, and consequently the crossover and mutation operators of genetic algorithm (GA) based on union-find sets is proposed. Thirdly, a self-adapting strategy that can adjust the acceleration coefficients is proposed to further improve the convergence and the quality of the proposed algorithm. Finally, the hybrid transformation can be applied to GA and the proposed algorithm can be applied to rectilinear architecture. To our best knowledge, the proposed algorithm is the first unified algorithm to solve the SMT construction under both octagonal and rectilinear architecture. The experimental results show that the proposed algorithm can efficiently provide a better solution for SMT problem both in octagonal and rectilinear architectures than others. Moreover, the algorithm can obtain several topologies of SMT, which is beneficial for optimizing congestion in VLSI global routing stage. [ABSTRACT FROM AUTHOR]

This paper addresses the multi-objective optimization of water–rail–road (WRR) intermodal transport system under uncertainty by explicitly capturing intermodal hub operation activities. Through the use of hub-and-spoke-type network, we formulate an uncertain multi-objective programming model for the WRR intermodal transportation network design problem, in which the cost, time and reliability objectives are simultaneously considered. Subsequently, we turn the original model into a deterministic equivalent multi-objective programming model under mild assumptions. Eventually, we utilize the ε -constraint method to reformulate the crisp multi-objective programming model to a modified mono objective one, which has proven to be NP-hard. Hence, we develop a memetic algorithm (MA) by combining a genetic algorithm and local intensification to solve the proposed problem. When designing the MA, we propose a combination encoding scheme to represent the location of intermodal hubs, the allocation of the demand nodes and the assignment of transportation modes. Moreover, we provide two local intensification operators to enhance exploitation ability. Finally, we implement a series of numerical experiments based on the Turkish network data set to verify the practicability of the proposed model and effectiveness of the solution approach developed in the paper. [ABSTRACT FROM AUTHOR]

In this article, a novel hybrid multi-objective bacterial colony chemotaxis (HMOBCC) algorithm is proposed to solve multi-objective optimization problems. A mechanism of particle swarm optimization is introduced to multi-objective bacterial colony chemotaxis (MOBCC) algorithm to improve the performance of MOBCC algorithm. Also, three other techniques, including dynamic reverse learning operator, external archive multiplying operator and adaptive diversity maintenance operator, are further applied to improve the diversity and convergence of the algorithm. The proposed algorithm is validated using 12 benchmark problems, and three performance measures are implemented for 5 benchmark problems to compare its performance with existing popular algorithms such as MOBCC, multi-objective bacterial colony chemotaxis based on grid algorithm, non-dominated sorting genetic algorithm (NSGA-II) and multi-objective evolutionary algorithm based on decomposition. The results show that the proposed HMOBCC is very effective against existing algorithms. The graphical abstract of this study. [ABSTRACT FROM AUTHOR]

In this paper, we consider the problem of mesh router placement in a wireless mesh network (WMN). The latter is an emerging networking technology consisting of three kinds of nodes: mesh clients, mesh routers and gateways. Mesh routers form a backbone to forward data between client nodes and the external network. Therefore, the optimization of mesh routers positions strongly influences the performance of the WMN. Since this issue has already been proved as being computationally NP-hard to solve, the use of non-exact methods (such as heuristics and metaheuristics) is indispensable. In this sense, our current work consists to apply and adapt the electromagnetism-like mechanism (EM) metaheuristic to solve the router node placement issue. The idea is to consider a population of solutions encoded as particles subject to attractions and repulsions as in electromagnetic systems. Finally, we have evaluated our proposed approach by simulating different scenarios under various settings. The obtained results indicate that the proposed EM algorithm outperforms the existing particle swarm intelligence algorithm and genetic algorithm in defining near optimal positions for mesh routers with regard to coverage and connectivity. [ABSTRACT FROM AUTHOR]

Smart metering in electricity markets offers an opportunity to explore more diverse tariff structures. In this article residential electricity demand and the System Marginal Price of Ireland's Single Electricity Market are simulated to estimate the wholesale risk associated with possible tariffs. A genetic algorithm (GA) with a stochastic fitness function is proposed to search for time-of-use tariffs that minimise wholesale risk to the supplier in residential markets. Alternative search algorithms and fitness functions are investigated in detail, as well as trade-offs in GA and simulation parameter settings. [ABSTRACT FROM AUTHOR]

Power distribution systems (PDS) in urban areas suffer from different types of problems. One such major problem is accidental or scheduled interruption. In electrical networks, effects of interruptions are usually quantified using a set of reliability indices, namely the System Average Interruption Frequency Index and the System Average Interruption Duration Index. Installation cost (fixed cost) and cost due to temporary and/or permanent faults during interruptions (variable cost) are also major issues to be considered while achieving a cost efficient, fault-tolerant PDS. Formalization of an optimization problem that jointly minimizes the afore-mentioned reliability indices as well as the cost of a PDS by optimal allocation of different protective devices and switches has always been a challenging task. This paper presents a hybrid single as well as joint-objective function optimization technique to minimize different reliability indices (mixed integer minimization problems), as well as the operational cost of a PDS in urban areas. In the proposed technique, two well-known meta-heuristic search techniques, namely genetic algorithms (GA) and ant colony optimization (ACO), have been hybridized after modifying different participating operators. The effectiveness of the proposed algorithm is examined, and each PDS is tested in a different environment of constrained optimization. In addition, the presented simulation results are compared with existing approaches that solve this problem. The simulation results show the superiority of the proposed hybrid GA-ACO model, as compared to other established heuristic approaches. [ABSTRACT FROM AUTHOR]

In multi-objective evolutionary optimization, maintaining a good balance between convergence and diversity is particularly crucial to decision makers, especially when tackling problems with complicated Pareto sets. According to the analysis of dominance-based and decomposition-based selection mechanisms in multi-objective evolutionary algorithms, a multi-objective evolutionary algorithm based on the combination of local non-dominated rank and global decomposition is presented. The Gauss distribution model and differential evolution based on history information are employed as evolutionary operators. Various comparative experiments are conducted on 19 unconstraint test MOPs, and our empirical results validate the effectiveness and competitiveness of our proposed algorithm in solving MOPs of different types. [ABSTRACT FROM AUTHOR]

In this manuscript, we identify and evaluate some of the most used optimization models for rule extraction using genetic programming-based algorithms. Six different models, which combine the most common fitness functions, were tested. These functions employ well-known metrics such as support, confidence, sensitivity, specificity, and accuracy. The models were then applied in the assessment of the performance of a single algorithm in several real classification problems. Results were compared using two different criteria: accuracy and sensitivity/specificity. This comparison, which was supported by statistical analysis, pointed out that the use of the product of sensitivity and specificity provides a more realistic estimation of classifier performance. It was also shown that the accuracy metric can make the classifier biased, especially in unbalanced databases. [ABSTRACT FROM AUTHOR]

Differential evolution (DE) is one of the most powerful and versatile evolutionary algorithms for efficiently solving complex real-world optimization problems in recent years. Since its introduction in 1995, the research focus in DE has mostly been on the variant side with so many new algorithms proposed based on the original DE algorithm. However, each new algorithm is only suitable for certain fitness landscapes, and, therefore, some types of optimization problems cannot be solved efficiently. To tackle this issue, this paper presents a new self-feedback DE algorithm, named the SFDE; its optimal variation strategy is selected by extracting the local fitness landscape characteristics in each generation population and combing the probability distributions of unimodality and multimodality in each local fitness landscape. The proposed algorithm is tested on a suite of 17 benchmark functions, and the experimental results demonstrated its advantages in a high search dimension in that it can ensure that the population moves to a better fitness landscape, then speeds up convergence to the global optimum, and avoids falling into local optima. [ABSTRACT FROM AUTHOR]

Inherent part of evolutionary algorithms that are based on Darwin’s theory of evolution and Mendel’s theory of genetic heritage, are random processes since genetic algorithms and evolutionary strategies are used. In this paper, we present extended experiments (of our previous) of selected evolutionary algorithms and test functions showing whether random processes really are needed in evolutionary algorithms. In our experiments we used differential evolution and SOMA algorithms with functions 2ndDeJong, Ackley, Griewangk, Rastrigin, SineWave and StretchedSineWave. We use n periodical deterministic processes (based on deterministic chaos principles) instead of pseudo-random number generators (PRGNs) and compare performance of evolutionary algorithms powered by those processes and by PRGNs. Results presented here are numerical demonstrations rather than mathematical proofs. We propose the hypothesis that a certain class of deterministic processes can be used instead of PRGNs without lowering the performance of evolutionary algorithms. [ABSTRACT FROM AUTHOR]

This paper addresses problem of predicting direction and magnitude of movement of currency pairs in the foreign exchange market. The study uses Support Vector Machine with a novel approach for input data and trading strategy. The input data contain technical indicators generated from currency price data (i.e., open, high, low and close prices) and representation of these technical indicators as trend deterministic signals. The input data are also dynamically adapted to each trading day with genetic algorithm. The study incorporates a currency strength-biased trading strategy which selects the best pair to trade from the available set of currencies and is an improvement over the previous work. The accuracy of the prediction models are tested across several different sets of technical indicators and currency pair sets, spanning 5 years of historical data from 2010 to 2015. The experimental results suggest that using trend deterministic technical indicator signals mixed with raw data improves overall performance and dynamically adapting the input data to each trading period results in increased profits. Results also show that using a strength-biased trading strategy among a set of currency pair increases the overall prediction accuracy and profits of the models. [ABSTRACT FROM AUTHOR]

Particle swarm optimization (PSO) algorithm has shown promising performances on various benchmark functions and engineering optimization problems. However, it is still difficult to achieve a satisfying trade-off between exploration and exploitation for all the optimization problems and different evolving stages. Furthermore, control parameters of some related mechanisms need pre-experience by the requirement of trial-and-error scheme. This paper presents a novel PSO algorithm, which adaptively adopts various search strategies, called Self-adapting Hybrid Strategy PSO (SaHSPS). Unlike some other peer PSO variants, this method dynamically changes the probabilities of different strategies according to their previous successful searching memories, without any additional control parameters. The probabilities of different strategies would be re-initialized according to a proposed dynamic probabilistic model to diverse the population. Besides, particles are updated by probabilistically selected strategies after niching PSO with Ring topology. Moreover, a dynamic updating mechanism by niching PSO is proposed to guarantee the parallel searching capability during the whole evolution process. Thus, this proposed algorithm might be problem-independent and search-stage-independent, yielding more satisfying solutions on various optimization problems. A comprehensive experimental study is conducted on 28 benchmark functions of CEC 2013 special session on real-parameter optimization, including shifted, rotated, multi-modal, high conditioned, expanded and composition problems, compared with several state-of-the-art variants of PSO and differential evolution (DE) algorithms. Comparison results show that SaHSPS obtains outstanding performances on the majority of the test problems. Moreover, a practical engineering problem, real power loss minimization of IEEE 30-bus power system, is used to further evaluate SaHSPS. The numerical results, compared with other stochastic search algorithms, show that SaHSPS could find high-quality solutions with higher probability. [ABSTRACT FROM AUTHOR]

In recent times, wireless access technology is becoming increasingly commonplace due to the ease of operation and installation of untethered wireless media. The design of wireless networking is challenging due to the highly dynamic environmental condition that makes parameter optimization a complex task. Due to the dynamic, and often unknown, operating conditions, modern wireless networking standards increasingly rely on machine learning and artificial intelligence algorithms. Genetic algorithms (GAs) provide a well-established framework for implementing artificial intelligence tasks such as classification, learning, and optimization. GAs are well known for their remarkable generality and versatility and have been applied in a wide variety of settings in wireless networks. In this paper, we provide a comprehensive survey of the applications of GAs in wireless networks. We provide both an exposition of common GA models and configuration and provide a broad-ranging survey of GA techniques in wireless networks. We also point out open research issues and define potential future work. While various surveys on GAs exist in the literature, our paper is the first paper, to the best of our knowledge, which focuses on their application in wireless networks. [ABSTRACT FROM AUTHOR]

Recently, how to reduce huge energy consumption of data centers has caught wide attention in cloud computing. One effective way is to improve the energy efficiency of servers. To achieve this goal, we propose a new energy-aware multi-job scheduling model based on MapReduce in this paper. In the proposed model, first, the variation of energy consumption with the performance of servers is taken into account; second, since network bandwidth is a relatively limited resource in cloud computing, 100 % data locality is guaranteed; last but not least, considering that task-scheduling strategies depend directly on data placement policies, we formulate the problem as an integer bi-level programming model. It is worth noticing that there are usually tens of thousands of tasks to be scheduled in the cloud, so this is a large-scale optimization problem. In order to solve it efficiently, a local search operator is specifically designed, based on which, a bi-level genetic algorithm is proposed in this paper. Finally, numerical experiments indicate the effectiveness of the proposed model and algorithm. [ABSTRACT FROM AUTHOR]

K-Means users usually have to decide on the number of clusters and the initial state by themselves. Evolutionary K-Means (EKM), a hybrid algorithm of K-Means and genetic algorithm, solves the problem by choosing the two parameters automatically through partition evolution; however, the final partition obtained often doesn't meet users' expectations. As a solution to this problem, we suggest using background knowledge for enhancing clustering quality and propose a semi-supervised approach that incorporates instance level constraints into the objective function of EKM. Firstly, we define Constrained Silhouette Index (CS) for data instances, which decreases the silhouette index of the instance having violated constraints. Then, we present two weighted approaches to extend the influence of constraints beyond the level of instance for evaluating the quality of a cluster or a partition. To evaluate the performance of CS in guiding EKM algorithms, we combine the two types of CS with F-EAC algorithm, and get two constrained EKM algorithms, which are named as CEAC1 and CEAC2, respectively. Experimental results on two artificial datasets and eight UCI datasets suggest a few constraints are often powerful enough to improve the accuracy of labelling instances and choosing K, and more constraints may improve the performance even more. [ABSTRACT FROM AUTHOR]

Imbalanced problems are quite pervasive in many real-world applications. In imbalanced distributions, a class or some classes of data, called minority class(es), is/are under-represented compared to other classes. This skewness in the data underlying distribution causes many difficulties for typical machine learning algorithms. The notion becomes even more complicated when machine learning algorithms are to combat multi-class imbalanced problems. The presented solutions for tackling the issues arising from imbalanced distributions, generally fall into two main categories: data-oriented methods and model-based algorithms. Focusing on the latter, this paper suggests an elegant blend of boosting and over-sampling paradigms, which is called MDOBoost, to bring considerable benefits to the learning ability of multi-class imbalanced data sets. The over-sampling technique introduced and adopted in this paper, Mahalanobis distance-based over-sampling technique (MDO in short), is delicately incorporated into boosting algorithm. In fact, the minority classes are over-sampled via MDO technique in such a way that they almost preserve the original minority class characteristics. MDO, in comparison with the popular method in this field, SMOTE, generates more similar minority class examples to original class samples. Moreover, the broader representation of minority class examples is provided via MDO, and this, in turn, causes the classifier to build larger decision regions. MDOBoost increases the generalization ability of a classifier, since it indicates better results with pruned version of C4.5 classifier; unlike other over-sampling/boosting procedures, which have difficulties with pruned version of C4.5. MDOBoost is applied to real-world multi-class imbalanced benchmarks and its performance is then compared with several data-level and model-based algorithms. The empirical results and theoretical analyses reveal that MDOBoost offers superior advantages compared to popular class decomposition and over-sampling techniques in terms of MAUC, G-mean, and minority class recall. [ABSTRACT FROM AUTHOR]

Invasive weed optimization (IWO) is a recent meta-heuristic optimization technique, based on the life cycle of plants. It has been applied in many engineering applications as well as in real world problems. In this paper, a hybrid version of IWO with the quadratic approximation (QA) operator, referred as QAIWO, has been investigated to improve the convergence rate of IWO while obtaining optimal solution. Additionally, we alleviate the limitation of QA (which is nothing but difficulty in escaping from a local optimum) by performing QA a predetermined number of times and then considering the average of all such solutions due to each iteration rather than a single solution. This technique makes our algorithm more efficient compared to the existing algorithms in the area. Twenty two benchmark problems and five real-life problems are adopted from literature to validate our proposed hybrid method QAIWO. The results of QAIWO are compared with the results obtained by the standard IWO and the well-known nature-inspired genetic algorithm (GA). These comparisons exhibit that QAIWO is more convenient to solve complex problems than using IWO and/or GA. [ABSTRACT FROM AUTHOR]

Instance selection is an important research problem of data pre-processing in the data mining field. The aim of instance selection is to reduce the data size by filtering out noisy data, which may degrade the mining performance, from a given dataset. Genetic algorithms have presented an effective instance selection approach to improve the performance of data mining algorithms. However, current approaches only pursue the simplest evolutionary process based on the most reasonable and simplest rules. In this paper, we introduce a novel instance selection algorithm, namely a genetic-based biological algorithm (GBA). GBA fits a 'biological evolution' into the evolutionary process, where the most streamlined process also complies with the reasonable rules. In other words, after long-term evolution, organisms find the most efficient way to allocate resources and evolve. Consequently, we can closely simulate the natural evolution of an algorithm, such that the algorithm will be both efficient and effective. Our experiments are based on comparing GBA with five state-of-the-art algorithms over 50 different domain datasets from the UCI Machine Learning Repository. The experimental results demonstrate that GBA outperforms these baselines, providing the lowest classification error rate and the least storage requirement. Moreover, GBA is very computational efficient, which only requires slightly larger computational cost than GA. [ABSTRACT FROM AUTHOR]

Rough set theory has been proven to be an effective tool to feature subset selection. Current research usually employ hill-climbing as search strategy to select feature subset. However, they are inadequate to find the optimal feature subset since no heuristic can guarantee optimality. Due to this, many researchers study stochastic methods. Since previous works of combination of genetic algorithm and rough set theory do not show competitive performance compared with some other stochastic methods, we propose a hybrid genetic algorithm for feature subset selection in this paper, called HGARSTAR. Different from previous works, HGARSTAR embeds a novel local search operation based on rough set theory to fine-tune the search. This aims to enhance GA's intensification ability. Moreover, all candidates (i.e. feature subsets) generated in evolutionary process are enforced to include core features to accelerate convergence. To verify the proposed algorithm, experiments are performed on some standard UCI datasets. Experimental results demonstrate the efficiency of our algorithm. [ABSTRACT FROM AUTHOR]

Dynamic optimization problems (DOPs) have attracted considerable attention due to the wide range of problems they can be applied to. Lots of efforts have been expended in modeling dynamic situations, proposing algorithms, and analyzing the results (too often in a visual way). Numeric performance measurements and their statistical validation have been however barely used in the literature. Most of works in DOPs report only the best-of-generation fitness, due to its simplicity of computation. Although this measure indicates the best algorithm in terms of fitness, it does not provide any details about the actual strength and weakness of each algorithm. In this article, we conduct a comparative study among algorithms of different search modes via several performance measures to demonstrate their relative advantages. We discuss the role of using different performance measures in drawing balanced conclusions about algorithms for DOPs. [ABSTRACT FROM AUTHOR]

Developing rule extraction algorithms from machine learning techniques such as artificial neural networks and support vector machines (SVMs), which are considered incomprehensible black-box models, is an important topic in current research. This study proposes a rule extraction algorithm from SVMs that uses a kernel-based clustering algorithm to integrate all support vectors and genetic algorithms into extracted rule sets. This study uses measurements of accuracy, sensitivity, specificity, coverage, fidelity and comprehensibility to evaluate the performance of the proposed method on the public credit screening data sets. Results indicate that the proposed method performs better than other rule extraction algorithms. Thus, the proposed algorithm is an essential analysis tool that can be effectively used in data mining fields. [ABSTRACT FROM AUTHOR]

The discretization of values plays a critical role in data mining and knowledge discovery. The representation of information through intervals is more concise and easier to understand at certain levels of knowledge than the representation by mean continuous values. In this paper, we propose a method for discretizing continuous attributes by means of fuzzy sets, which constitute a fuzzy partition of the domains of these attributes. This method carries out a fuzzy discretization of continuous attributes in two stages. A fuzzy decision tree is used in the first stage to propose an initial set of crisp intervals, while a genetic algorithm is used in the second stage to define the membership functions and the cardinality of the partitions. After defining the fuzzy partitions, we evaluate and compare them with previously existing ones in the literature. [ABSTRACT FROM AUTHOR]

Studies on parameter tuning in evolutionary algorithms are essential for achieving efficient adaptive searches. This paper discusses parameter tuning in real-valued crossover operators theoretically. The theoretical analysis is devoted to improving robustness of real-coded genetic algorithms (RCGAs) for finding optima near the boundaries of bounded search spaces, which can be found in most real-world applications. The proposed technique for crossover-parameter tuning is expressed mathematically, and thus enables us to control the dispersion of child distribution quantitatively. The universal applicability and effect have been confirmed theoretically and verified empirically with five crossover operators. Statistical properties of several practical RCGAs are also investigated numerically. Performance comparison with various parameter values has been conducted on test functions with the optima placed not only at the center but also in a corner of the search space. Although the parameter-tuning technique is fairly simple, the experimental results have shown the great effectiveness. [ABSTRACT FROM AUTHOR]