Search

Differential Evolution (DE) is a population-based metaheuristic search algorithm that exhibits excellent performance. However, it is sensitive to mutation strategies and control parameters.To mitigate the impact of these factors, this paper proposes refining differential evolution with mutation rate and neighborhood weight local search(MRNLDE). The algorithm guides individuals in selecting an appropriate mutation mode through the mutation rate, effectively utilizing the evolutionary information inherent in the individuals themselves, thereby enhancing the search efficiency of the algorithm. Furthermore, the neighborhood individual information is utilized to re-explore the solution space, allowing for a comprehensive exploration of individual evolutionary potential and enhancing the algorithm's diversity. The performance of MRNLDE was validated under two sets of benchmark problems at the Institute of Electrical and Electronics Engineers (IEEE) Conference on Computing in Evolution (CEC), and the results show that MRNLDE performs well overall. [ABSTRACT FROM AUTHOR]

This research proposes a novel Hybrid Success History Intelligent Optimizer with Gaussian Transformation (SHIOGT) for solving different complexity level optimization problems and for Convolutional Neural Network (CNNs) hyperparameter tuning. SHIOGT algorithm is designed to balance exploration and exploitation phases through the addition of Gaussian Transformation to the original Success History Intelligent Optimizer. The inclusion of Gaussian Transformation enhances solution diversity enables SHIO to avoid local optima. SHIOGT also demonstrates robustness and adaptability by dynamically adjusting its search strategy based on problem characteristics. Furthermore, the combination of Gaussian and SHIO facilitates faster convergence, accelerating the discovery of optimal or near-optimal solutions. Moreover, the hybridization of these two techniques brings a synergistic effect, enabling SHIOGT to overcome individual limitations and achieve superior performance in hyperparameter optimization tasks. SHIOGT was thoroughly assessed against an array of benchmark functions of varying complexities, demonstrating its ability to efficiently locate optimal or near-optimal solutions across different problem categories. Its robustness in tackling multimodal and deceptive landscapes and high-dimensional search spaces was particularly notable. SHIOGT has been benchmarked over 43 challenging optimization problems and have been compared with state-of-the art algorithm. Further, SHIOGT algorithm is applied to the domain of deep learning, with a case study focusing on hyperparameter tuning of CNNs. With the intelligent exploration–exploitation balance of SHIOGT, we hypothesized it could effectively optimize the CNN's hyperparameters. We evaluated the performance of SHIOGT across a variety of datasets, including MNIST, Fashion-MNIST, CIFAR-10, and CIFAR-100, with the aim of optimizing CNN model hyperparameters. The results show an impressive accuracy rate of 98% on the MNIST dataset. Similarly, the algorithm achieved a 92% accuracy rate on Fashion-MNIST, 76% on CIFAR-10, and 70% on CIFAR-100, underscoring its effectiveness across diverse datasets. [ABSTRACT FROM AUTHOR]

The cooperative coevolution framework was introduced to address the shortcomings of metaheuristic algorithms in solving continuous large-scale global optimization problems. By dividing the problem into subcomponents which can be optimized separately, the framework can improve on both the solution's quality as well as the computational speed by exposing a degree of parallelism. Distributed computing platforms, such as Apache Spark, have long been used to improve the speed of different algorithms in solving computational problems. This work proposes a distributed implementation of the cooperative coevolution framework for solving large-scale global optimization problems on the Apache Spark distributed computing platform. By using a formerly outlined distributed variant of the cooperative coevolution framework, features of the Spark platform are utilized to enhance the computational speed of the algorithm while maintaining comparable search quality to other works in the literature. To test for the proposed implementation's improvement in computational speed, the CEC 2010 large-scale global optimization benchmark functions are used due to the diversity they offer in terms of complexity, separability and modality. Results of the proposed distributed implementation suggest that a speedup of up to ×3.36 is possible on large-scale global optimization benchmarks using the Apache Spark platform. [ABSTRACT FROM AUTHOR]

Scheduling tasks in the cloud system is the main issue that needs to be addressed in order to improve customer satisfaction and system performance. This paper proposes DCOHHOTS, a novel multi-objective task scheduling algorithm based on a modified Harris hawks optimizer. In overall, this paper has two main stages. As the first step, DCOHHO is introduced as a new version of Harris Hawks Optimizer. Using the Differential Evolution algorithm, an optimal configuration is selected from the chaotic map, the opposition-based learning, and the ratio of the population. In order to improve the performance of the Harris Hawks Optimizer, this optimal configuration is applied to initialize the hawk's position. In the second stage, DCOHHOTS, a DCOHHO-based Task Scheduling algorithm, is proposed. Multi-objective behavior in the proposed task scheduling algorithm optimizes resource utilization to decrease the makespan, energy consumption, and execution cost. Moreover, prioritizing tasks before submitting them to the scheduler is done using the hierarchical process in the DCOHHOTS algorithm. For the purpose of investigating the performance of the proposed DCOHHO algorithm, a number of experiments are conducted using 20 standard functions and twelve algorithms. The experimental results demonstrate that the DCOHHO algorithm is superior at determining the optimal test function solutions. Additionally, makespan, execution cost, resource utilization, and energy efficiency of DCOHHOTS task scheduling algorithms are analyzed. Compared to existing algorithms, the proposed algorithm saves up to 16% energy in heavy loads. Additionally, resource utilization has increased by 17%. Compared to the conventional algorithm, the proposed algorithm reduced makepan and execution cost by 26% and 8%, respectively. [ABSTRACT FROM AUTHOR]

Benefiting from cloud computing's elasticity, scalability, and pay-per-use model, more and more scientific applications are deployed in or migrated to the cloud. Workflow scheduling still faces many challenges due to the growing scales of workflows and the diversified user QoS requirements. In this work, we propose a Mutation-driven and population Grouping Poor and Rich Optimization algorithm (MG-PRO) for scheduling workflows in the cloud to minimize makespan while satisfying the budget constraints. Specifically, we first adopt the middle-class sub-population into the original Poor and Rich Optimization algorithm (PRO), and develop the update strategies for rich and middle-class sub-populations to increase the randomness and search diversity. Secondly, the update mechanism for rich individuals is enriched, and the middle-class sub-population is guided by elite rich individuals, which enhances the information exchange and sharing among sub-populations. Finally, an evolution-aware mutation strategy is designed, where the mutation probability is adjusted adaptively as the dynamic monitoring of the population update process, and the two-point and triangular crossover-based mutations are used alternately to intervene the evolution trajectory according to the degree of objective optimization, resulting in a better balance between exploration and exploration. Extensive experiments are conducted on well-known scientific workflows with different types and scales through WorkflowSim. The experimental results show that, in most cases, MG-PRO outperforms existing algorithms in terms of constraint satisfiability, solution quality and stability. It can generate near-optimal solutions with the different budget constraints satisfied in a relatively short time, for example, the makespan resulting from MG-PRO is at most 59.95% shorter than other meta-heuristic algorithms, and at least 7.33% shorter than all its peers. [ABSTRACT FROM AUTHOR]

Extant sequential wrapper-based feature subset selection (FSS) algorithms are not scalable and yield poor performance when applied to big datasets. Hence, to circumvent these challenges, we propose parallel and distributed hybrid evolutionary algorithms (EAs) based wrappers under Apache Spark. We propose two hybrid EAs based on the Binary Differential Evolution (BDE), and Binary Threshold Accepting (BTA), namely, (i) Parallel Binary Differential Evolution and Threshold Accepting (PB-DETA), where BDE and BTA work in tandem in every iteration, and (ii) its ablation variant, Parallel Binary Threshold Accepting and Differential Evolution (PB-TADE). Here, BTA is invoked to enhance the search capability and avoid premature convergence of BDE. For comparison purposes, we also parallelized two state-of-the-art algorithms: adaptive DE (ADE) and permutation based DE (DE-FSPM), and named them PB-ADE and P-DE-FSPM respectively. Throughout, logistic regression (LR) is employed to compute the fitness function, namely, area under the receiver operator characteristic curve (AUC). The effectiveness of the proposed algorithms is tested over the five big datasets of varying dimensions. It is noteworthy that the PB-TADE turned out to be statistically significant than the rest. All the algorithms have shown the repeatability property. The proposed parallel model attained a speedup of 2.2–2.9. We also reported feature subset with high AUC and least cardinality. [ABSTRACT FROM AUTHOR]

System virtualization is a key technology nowadays for reducing energy consumption and investment in server computers. However this environment frequently faces the problem of virtual machine placement (VMP). In order to solve this problem effectively, we have proposed a Dual-Operator-based Constrained many-objective Evolutionary Algorithm, DOCEA/D, which employs a Decomposition-based strategy. The algorithm uses differential evolution (DE) as an evolutionary mechanism and also employs a novel diversity maintenance (DM) technique to avoid trapping in local optima. For the purpose of validating the performance of DOCEA/D, it has been compared here with other contemporary many-objective evolutionary algorithms (MaOEAs) namely MOEA/D, NSGA-III and IBEA on three instances of VMP problem. The results on the different quality indicators clearly demonstrate the superiority of DOCEA/D over the counterpart approaches. As a further proof, we graphically show the superiority of our proposed method over the competing approaches and also demonstrate the statistical significance of our results. [ABSTRACT FROM AUTHOR]

Differential Evolution (DE) is perhaps one of the most efficient and versatile evolutionary techniques and its different variants have been proposed in the recent past. However, all these variants often suffer from the shortcomings of slow and/or premature convergence. Further, nearest neighbour (NN) is a frequently used classifier in the domain of data mining but its primitive version is noise sensitive and shows low accuracy. Different data reduction methods such as prototype selection, prototype generation, and DE are widely used to mitigate these drawbacks of NN. However, none of the methods substantially avoids all these shortcomings. This paper proposes a new variant of DE that is based on the biological phenomenon of the centromeric location of chromosomes. The proposed 'Chromosomes Identification based Differential Evolution (CIDE)' extracts the knowledge from databases and constructs effective and reliable decision rules for classification. In this paper, the performance of the proposed CIDE technique is computed and compared with 1-NN classifier optimized by fourteen state-of-art variants of DE, and 1-R classifier using six-real datasets as well as KDDCup'99 dataset to illustrate its superiority over the 1-NN and 1-R classifier. Finally, the performance of CIDE is also compared with Naïve Bayes, support vector machine, Neural Network, and Random Forest for the KDDCup'99 dataset using different performance metrics. [ABSTRACT FROM AUTHOR]

Travelling Salesman Problem (TSP) is an Np-Hard problem, for which various solutions have been offered so far. Using the Harris Hawk Optimization (HHO) algorithm, this paper presented a new method that uses random-key encoding to generate a tour. This method helps maintain the main capabilities of the HHO algorithm, on the one hand, and to take advantage of the capabilities of active mechanisms in the continuous-valued problem space on the other hand. For the exploration phase, the DE/best/2 mutation mechanism employed in the exploitation phase, besides the main strategies in the HHO algorithm, was used. Ten neighborhood search operators are used, four of which are introduced. These operators were intelligently selected using the MCF. The Lin-Kernighan local search mechanism was utilized to improve the proposed algorithm's performance, and the Metropolis acceptance strategy was employed to escape the local optima trap. Besides, 80 datasets were evaluated in TSPLIB to demonstrate the performance and efficiency of the proposed algorithm. The results showed the excellent performance of the proposed algorithm. [ABSTRACT FROM AUTHOR]

With the rapid advancements in processing and storage technology along with the popularity of the internet, computing capabilities have become more affordable, efficient, and widely accessible than ever before. This advancement has resulted in the emergence of a modern computing environment known as fog computing. Due to the latency-sensitiveness feature, computation of these services in fog computing is advantageous than cloud. Task scheduling is a significant issue in fog systems and substantially impacts resource utilization, task computation, and latency time. Many heuristic and meta-heuristic techniques have been applied to solve the scheduling issue. For the success of any meta-heuristic algorithm, an appropriate composition of exploration and exploitation of solution space is required to improve convergence and avoid local optima. To meet these requirements, a modified fireworks algorithm with the combination of opposition-based learning and differential evolution techniques is presented. Differential evolution operator has been utilized to avoid local optima and opposition-based learning technique has been applied for creating a diversified solution set of population. The proposed method works on the minimization of makespan and cost and improves resource utilization. The experiments have been carried out on a variety of workloads, and the findings have been compared with some recent popular metaheuristic techniques. The comparison has verified the importance of the proposed approach. [ABSTRACT FROM AUTHOR]

Based on the algorithm structure, each metaheuristic algorithm may have its pros and cons, which may result in high performance in some problems and low functionality in some others. The idea is to hybridize two or more algorithms to cover each other's weaknesses. In this study, particle swarm optimization (PSO), simulated annealing (SA) and differential evolution (DE) are combined to develop a more powerful search algorithm. First, the temperature concept of SA is applied to balance the exploration/exploitation capability of the hybridized algorithm. Then, the DE's mutation operator is used to improve the exploration capability of the algorithm to escape the local minimums. Next, DE's mutation operator has been modified so that past experiences can be used for smarter mutations. Finally, the PSO particles' tendency to their local optimums or the global optimum, which balances the algorithm's random and greedy search, is affected by the temperature. The temperature influences the algorithm's behavior so that the random search is more significant at the beginning, and the greedy search becomes more important as the temperature is reduced. The results are compared with the basic PSO, SA, DE, cuckoo search (CS), and hybridized CS-PSO algorithm on 20 benchmark problems. The comparison reveals that, in most cases, the new algorithm outperforms others. [ABSTRACT FROM AUTHOR]

With the rapid increase in the use of cloud computing systems, an efficient task scheduling policy, which deals with the assignment of tasks to resources, is required to obtain maximum performance. Cloud task scheduling (CTS) is an established NP-Hard optimization problem that can be effectively tackled with meta-heuristic algorithms. The cuckoo search (CS) algorithm is a powerful swarm-intelligence meta-heuristic that has been successfully applied over a wide-range of real-life optimization problems, including task scheduling problems. Besides its strong exploration ability, the CS algorithm suffers from insufficient local search, lack of solution diversity towards the end, and slow convergence problem. These drawbacks produce inefficient cloud task schedules resulting in sub-optimal performance. In this manuscript, an improved CS-based scheduling algorithm called CSDEO is introduced, which combines the features of the Opposition-based learning (OBL) method, Cuckoo search, and Differential evolution (DE) algorithms to optimize workload makespan and energy consumption of the cloud resources. Our CSDEO algorithm firstly uses the OBL method to produce an optimal initial population by providing solutions across the entire solution space. Then, the CSDEO uses an effective way of switching between the CS exploration phase and the DE exploitation phase, depending on each solution's fitness. Experiments are conducted on the CloudSim simulator by using the CEA-Curie and HPC2N supercomputing workloads. The observations show that in the case of CEA-Curie workloads, the proposed CSDEO algorithm achieves makespan improvement in the range of 6.29–29.76% and energy consumption improvement in the range of 3.76–201.98% over well-known scheduling algorithms. In the case of HPC2N workloads, the improvement ranges of the CSDEO approach for the makespan and energy consumption metrics are 9.86–281.69% and 6.12–233.3%, respectively compared to the tested scheduling algorithms. [ABSTRACT FROM AUTHOR]

Over the past few years, cloud computing model (e.g., Spark) has aroused huge attention. Differential evolution (DE) has been applied to cloud computing models by a number of researchers for its merits in solving large-scale global optimization problems (LSGO), and remarkable results have been achieved. Moreover, we noticed that a combination of better topology and migration strategy is critical to solve the mentioned problems when DE algorithm acts as an internal optimizer for Spark cloud computing model. However, rare studies have been conducted to combine the combination to enhance the performance of DE algorithm for solving large-scale global optimization problems. Thus, inspired by the mentioned insights, we propose a novel grouping topology model that uses DE variants as internal optimizers to solve LSGO problems, called SgtDE. In SgtDE, population is split into subgroups evenly, and various topology structures are introduced to migrate individuals between and within subgroups. In this paper, five types of DE are adopted as the internal optimizers. By comparing the 20 benchmark functions presented on CEC2010, the results demonstrate that the SgtDE, especially a combination of better topology and migration strategy, exhibits significant performance in applying various DE variants. Thus, the SgtDE can act as the next generation optimizer of the cloud computing platform. [ABSTRACT FROM AUTHOR]

Efficient task scheduling is considered as one of the main critical challenges in cloud computing. Task scheduling is an NP-complete problem, so finding the best solution is challenging, particularly for large task sizes. In the cloud computing environment, several tasks may need to be efficiently scheduled on various virtual machines by minimizing makespan and simultaneously maximizing resource utilization. We present a novel hybrid antlion optimization algorithm with elite-based differential evolution for solving multi-objective task scheduling problems in cloud computing environments. In the proposed method, which we refer to as MALO, the multi-objective nature of the problem derives from the need to simultaneously minimize makespan while maximizing resource utilization. The antlion optimization algorithm was enhanced by utilizing elite-based differential evolution as a local search technique to improve its exploitation ability and to avoid getting trapped in local optima. Two experimental series were conducted on synthetic and real trace datasets using the CloudSim tool kit. The results revealed that MALO outperformed other well-known optimization algorithms. MALO converged faster than the other approaches for larger search spaces, making it suitable for large scheduling problems. Finally, the results were analyzed using statistical t-tests, which showed that MALO obtained a significant improvement in the results. [ABSTRACT FROM AUTHOR]

In this paper, we consider the contention aware task scheduling problem on a grid topology of processors. By contention awareness, we mean that simultaneous communication on a link has to be serialized. To solve this problem, we propose several nature inspired metaheuristic algorithms: Simulated Annealing (SA), Genetic Algorithm (GA), Differential Evolution (DE), Particle Swarm Optimization (PSO), Bat Algorithm (BA), Cuckoo Search (CS), and Firefly Algorithm (FA). We perform benchmark evaluation of these algorithms for the Normalized Schedule Length (NSL) parameter. The benchmark task graphs that we consider are: random task graphs, peer set task graphs, systolic array task graphs, Gaussian elimination task graphs, divide and conquer task graphs, and fast Fourier transform task graphs. [ABSTRACT FROM AUTHOR]

Delivering relief supplies to victims of natural disasters is a very complicated process faced with many challenges: damaged road network, high demand for various materials, scarcity of resources, multi-supply and multi-demand sites, limited transport capacity, etc. Traditionally, outreach maximization is the focus of emergency relief. In this research, we propose a bi-level programming model which takes the aforementioned challenges into full account. We consider two objectives: minimizing the distribution time, and maximizing the allocation fairness. Because the problem is NP-hard, we design an improved differential evolution (IDE) algorithm and numerically compare it with several conventional differential evolution algorithms, including CoDE, SaDE, JADE and JDE. We found that the proposed IDE outperforms the existing algorithms. The feasibility and validity of the proposed model and the IDE algorithm are verified by applying them to the 2008 Wenchuan earthquake in western China. [ABSTRACT FROM AUTHOR]

Based on the evolutionary algorithms of the four complex networks, the evolution of knowledge network is regarded as that of complex networks. With the heterogeneity of knowledge level, knowledge absorptive and innovative capacity and agents' knowledge types considered, theoretical models of knowledge network evolution are constructed. Through numerical simulation, different network structures are analyzed in terms of their effects on the diffusion efficiency of the overall knowledge as well as of various types of knowledge. The simulation results show that: with the diffusion of the overall knowledge considered, although the overall knowledge level in a small-world structure is lower than the random network in the early and middle stage, it is close to the highest one later on; moreover, its growth rate is relatively higher among all four networks and its knowledge levels are distributed most uniformly. With regard to the diffusion of different types of knowledge, the small-world network is proved to produce the most uniform gap between knowledge types and help those dominant industries in the early stage remain advanced during the evolutionary process. [ABSTRACT FROM AUTHOR]

In this paper, a discrete differential algorithm is proposed to extract data features. The strategy of regional tourism development, features extraction and the technology of retaining the subject information of industry characteristic data are analyzed to reduce the feature space dimension and achieve data dimension reduction through the key factors in the development of tourism industry in the region. In this paper, a kind of distributed feature selection system is obtained based on local fisher discriminant according to the feature selection method based on data structure and the feature of modal distribution in industrial characteristic data. The experimental results show that this feature can make the space describe the data distribution structure accurately after dimension reduction. [ABSTRACT FROM AUTHOR]

Optimal placement of phasor measurement unit (OPPMU) considering controlled islanding is a high dimensional constrained optimization problem that determines optimum location of PMU. Controlled islanding can be used as an effective emergency action to split a large scale power system into islands for avoiding blackouts. This paper proposes an optimal PMU placement model considering power system controlled islanding. So that the power network remains observable under controlled islanding as well as normal operating condition. A new method based on spectral clustering is proposed to handle multiple constraints in order to guarantee the stability of generated islands. The objective function used in this controlled islanding algorithm is the minimal power flow disruption. Constraints such as generator coherency, load–generation imbalance are considered for spectral clustering. OPPMU is formed for the obtained islands using two conflicting objectives of minimizing the number of PMU and maximizing the measurement redundancy using conventional mixed integer linear programming, real coded genetic algorithm(RGA), non dominated sorting differential evolutionary algorithm, and modified non dominated sorting GA II (MNSGA II). RGA with simulated binary crossover and non dominated sorting differential evolution with improved crowding distance and mutation is implemented. In MNSGA II proposed a variant which reduces the run-time complexity using the technique space–time-trade-off. Simulation results using NE 39 and IEEE 118 bus test systems show that the proposed method is computationally efficient when solving controlled islanding based OPPMU problem. [ABSTRACT FROM AUTHOR]

As one of the major problems is that the time taken for executing the traditional algorithm is larger and that it is very difficult for processing large amount of data. Clusters possess high degree of similarity among each cluster and have low degree of similarity among other clusters. Optimization algorithm for clustering is the art of allocating scarce resources to the best possible effect. The traditional optimization algorithm is not suitable for processing high dimensional data. The main objective of proposed Artificial Bee Colony (ABC) approach is to minimize the execution time and to optimize the best cluster for the various sizes of the dataset. To deal with this, we are normalizing to distributed environment for time efficiency and accuracy. The proposed ABC algorithm simulates the behavior of real bees for solving numerical optimization problems particularly in clustering. The dataset size is varied for the algorithm and is mapped with its appropriate timings. The result is observed for various fitness and probability value which is obtained from the employed and the onlooker phase of ABC algorithm from which the further calibrations of classification error percentage is done. The proposed ABC Algorithm is implemented in Hadoop environment using mapper and reducer programming. An experimental result reveals that the proposed ABC scheme reduces the execution time and classification error for selecting optimal clusters. The results show that the proposed ABC scheme gives a better performance than PSO and DE in terms of time efficiency. [ABSTRACT FROM AUTHOR]

Scalable computing resources are provided via the Internet in the cloud computing environment. A growing number of application providers begin to deploy their applications in cloud to save the infrastructure maintaince cost. The probability of node failures cannot be nontrivial due to a great quantity of nodes in the cloud data center. To address the problem, the virtual machine replication technique is extensively adopted in the cloud system to enhance the application/service reliability. K-fault tolerance is a typical replication strategy employed in cloud. However, currently proposed K-fault tolerance replication strategies cannot achieve the best effect due to the ignorance of switch failure. In this paper, we study to design a (m, n)-fault tolerance virtual machine placement algorithm to solve the problem. Firstly, we formulate the problem as an integer linear programming problem, and prove that the problem is NP-hard. Secondly, we extensively employ differential evolution (DE) algorithm to solve the integer linear programming problem. Finally, experiments are conducted to study the effectiveness of our algorithm, and the simulation results demonstrate that our algorithm outperforms other algorithms in reliability enhancement. [ABSTRACT FROM AUTHOR]

Control parameters and mutation methods impact upon the global search ability of differential evolution algorithm (DE), and varying optimization issues own varying parameter settings. In this paper, an enhanced elite archive mutation strategy with self-adaption parameter adjustment (EAMSADE) is proposed to raise DE's performance. The population's diversity and the individual's difference are considered by this paper to enhance the algorithm's convergence property. EAMSADE amends the DE/rand/1 strategy by means of enhanced elite archive mutation and modifies parameters (crossover rate and scaling factor) adaptively which is based on quantitative analysis of individual variability and population diversity. To confirm the proposed EAMSADE's performance, a suit of 21 benchmark functions from IEEE CEC2005 are utilized to carry out the experiment. The outcome of the experiment confirms that the proposed EAMSADE has got an overall improvement on convergence performance and global search ability compared to the other four amended DE. [ABSTRACT FROM AUTHOR]

The biggest feature of the era of big data is that people can easily generate, access, and make use of massive data resources. As one of the most important and popular kind of big data, location big data and its application technology provide users with convenient services. However, improper collection, analysis and publishing of location big data also brings huge crisis of personal privacy disclosure. Spatial decomposition is one of the effective ways to achieve the statistics publication of location big data. In order to make full use of the redundant characteristics of location big data in spatial and temporal distribution, a hierarchical differential privacy hybrid decomposition algorithm is proposed in this paper. In the first layer of decomposition, an adaptive density grid structure is used to cluster the location big data, which not only reduces the uniform assumption errors but also avoids noise errors caused by large number of empty nodes. In order to guide the reasonable decomposition for skewed grids in the second layer, a heuristic quad-tree decomposition algorithm based on regional uniformity is designed, which solved the difficult problem for determining stop condition of the top-down decomposition of two-dimensional space. Comparative experiments show that the hierarchical differential privacy hybrid decomposition algorithm proposed in this paper has good effect in improving the accuracy of regional counting queries. The proposed algorithm has low computational complexity and obvious advantages in the publishing environment of big data. [ABSTRACT FROM AUTHOR]

In mobile wireless sensor networks (MWSNs), methods of traditional data collection only consider increasing the amount of data acquisition or reducing the energy consumption of the whole network. In the process of data collection, how to maximize the data, the shortest mobile path and the reliability of the network is an optimization problem. In order to solve the above problems, this paper proposes an algorithm of reliable data collection for mobile Sink based on improved artificial bee colony optimization. It combines the selection of cluster node, the transmission path from sensor node to cluster node, and the path optimization of mobile Sink. It provides a typical model system of data collection in MWSNs, including the method of network energy consumption in the process of data collection. An improved artificial bee colony algorithm is proposed by using the initialization method of reverse learning, and introducing the search equation inspired by differential evolution algorithm. It overcomes the shortcomings of artificial bee colony algorithm with premature convergence and poor ability to search in late evolution, which satisfying the conditions of the shortest time consumption and the shortest path length of the mobile Sink. On the one hand, it aims to improve data collection and find the shortest traversal path of the mobile Sink. On the other hand, it is necessary to improve the network efficiency and reliability by considering network energy consumption and network delay. By the improved artificial bee colony algorithm, it gets the shortest path planning of mobile Sink for searching each cluster node. The sensor node transmits data to the nearest cluster node by multi hop routing with temporarily saving, and then it is sent to the mobile Sink. The proposed algorithm can effectively reduce the amount of sensor nodes transmitted to mobile sink with improving the efficiency of data collection. Compared with other methods, it can reduce network energy consumption and increase energy consumption balance and network reliability, so as to prolong network lifetime. [ABSTRACT FROM AUTHOR]

Dynamic economic dispatch optimum scheduling of power plant generation is of great importance to electric utility systems, it is difficult to solve because of its complex structure, variable parameter, nonlinear characteristics et al. Based on analysis of DE searching mechanism, an improved differential evolution (IDE) algorithm based DE/target-to-best is presented,which adopts an improved mutation strategy that a random vector and the previous best vector is used instead of the current vector in case the DE algorithm may be in early maturity or decline in convergence speed.The algorithm is applied to solve the generators dynamic load economic dispatch problems taking into account the incremental fuel cost function and the valve-point effects. Computer simulation test shows that IDE algorithm provides better solution of less cost In the case of less generations and outperforms GA, PSO and DE. [ABSTRACT FROM AUTHOR]

In order to improve the ability of resource allocation and scheduling in cloud computing, optimize resource allocation and improve the efficiency of cloud computing, an optimal computing resource allocation algorithm in cloud computing based on hybrid differential parallel scheduling is proposed. In this algorithm, the models of data structure and gird structure of computing resource allocation in cloud computing are constructed, and the sample clustering analysis method of resource information flow is used to classify the attributes of computing resources; the sliding window of computing resource allocation is divided into multiple sub-windows; characteristic quantities associated with computing resource allocation attributes are selected in neighbor samples as standard vector sets for adaptive pairing; the computing resources in cloud computing are done with singular value decomposition and the resource allocation is transformed into the least square problem; the hybrid differential parallel computing method is used for optimal solution finding of resource scheduling vector set to prevent the allocation results from falling into local optimal solution, so as to improve the global convergence of resource allocation. The simulation results show that when the method proposed in this paper is used for resource allocation in clouding computing, the clustering performance is high and the convergence control ability to computing resources with different attributes is high; the allocation speedup can reach 3.67, which is improved by 14.65 and 7.43% respectively compared with that in the traditional HEFT algorithm and HCNF algorithm; when the number of allocate nodes is 100, the overhead is only 5.6, which is reduced by 14.56 and 8.33% than that in traditional HEFT algorithm and HCNF algorithm. So it shows that the proposed method has a higher practical application performance for its shorter execution time and lower overhead. [ABSTRACT FROM AUTHOR]

The imbalanced datasets are common in real-world application and the problem of imbalanced dataset affect classification performance of many standard learning approaches. To address imbalanced datasets, a weighted extreme learning machine (WELM) solving the L 2 -regularized weighted least squares problem is presented to avoid the generation of an over-fitting model and obtain better generalization ability compared with ELM. However, the weight generated according to class distribution of training data leads to lack of finding optimal weight with good generalization performance and the randomness of input weight and hidden biases of network makes the algorithm produce suboptimal classification model. In this paper, a weighted extreme learning machine based on hybrid artificial bee colony (HABC) is proposed to obtain better performance than WELM, in which input weights and hidden bias of WELM and the weight assigned to training samples are optimized by the hybrid artificial bee colony algorithm. HABC combines the diversities of the perturbed parameter vectors of differential evolution with the best solution information of the artificial bee colony effectively. In the empirical study, different class imbalance data handling methods including four WELM-based methods, weighted support vector machine, four ensemble methods which combine data sampling and the Bagging or Boosting are compared with our method. The experimental results on 15 imbalanced datasets show that the proposed method outperforms most methods, which indicates its superiority. [ABSTRACT FROM AUTHOR]

More and more distributed generations are connected in the distribution network. Intermittent output and different locations have a significant impact to the distribution network voltage, current, power flow, the traditional forward and backward substitution is unable to solve PV-type node and meshed network, the distributed generations increase the number of network constraints and increase the difficulty of searching the optimal solution. To solve the problem of classic self-healing method failure, based on the multi-agent system, a novel self-healing reconfiguration scheme is proposed for distribution network with distributed generations in this study. Multiple objectives are considered for minimum distributed generation output loss, minimum power loss, load balancing among the feeders and branch current constraint violation, improved forward-backward weep method is used to get power flow solution for different node types of distributed generations, a self-adaptive differential evolution algorithm with improved strategies is proposed to solve problem. The performance of proposed algorithm is analyzed for several case studies on IEEE 33-bus system. The simulation results show that the approach can improve the self-healing reconfiguration performance and adapt to the changes of dynamic conditions. [ABSTRACT FROM AUTHOR]

A design method of external elevation of high-rise Building based on emotional differential evolution algorithm in urban landscape design was proposed in this thesis to realize automatic generation of layout of high-rise building in urban landscape design. Firstly, brief description was conducted on automatic generation process of external elevation layout for high-rise building in differential evolution algorithm and urban landscape design and analysis was conducted on principle of differential evolution algorithm. Meanwhile, an improved algorithm of differential evolution based on social emotional operation was proposed to increase optimizing and rapid convergence ability of algorithm to realize optimization of design method of external elevation for high-rise building. Experimental result indicates that proposed algorithm has better optimization performance and experimental analysis was conducted on application of the algorithm in external elevation design of high-rise building. [ABSTRACT FROM AUTHOR]

In order to improve the validity of comparative literature research method, a kind of research method for comparative literature based on neural network eco-criticism theory system is proposed. Firstly, assessment indicator system of comparative literature research based on eco-criticism theory is established by social benefit, economic benefit, influence and writer information and other indicators; secondly, factors proportion of ecology assessment indicator for comparative literature is carried out by introducing neural network, and at the same time, in order to improve the validity of neural network algorithm, the parameter optimization setting of neural network algorithm is achieved by designed differentiation Gauss double-orientated differential evolution algorithm; finally, the validity of proposed method is verified by simulation experiment in comparative literature research. [ABSTRACT FROM AUTHOR]

Recently, information security suited for a comprehensive number of practical applications and which mainly affected by an end-to-end security. Earlier works on information security have concentrated on protecting messages rather than aggregating whether the message reliability. In order to overcome these drawbacks, in this paper, a hybrid aggregated signcryption (HAS) scheme is proposed that uses the efficient key encapsulation mechanism (KEM) and data encapsulation mechanism (DEM). The KEM utilizes the modified version of Kurosawa and Desmedt hashing scheme to encapsulate the key. The DEM utilizes the elliptic curve cryptography (ECC) algorithm to encapsulate the message. The multi-constraints differential evolution (MDE) algorithm used to select optimal prime fields in ECC algorithm. Furthermore, an aggregated signature scheme used to aggregate many signatures on many messages from many distinct singers into one single signature, which minimizes the verification cost. The proposed HAS scheme achieves reliability and unforgeablity rather than existing schemes. [ABSTRACT FROM AUTHOR]

A cloud service provider (CP) offers computing resources with their own interface type and pricing policies besides other services such as storage on a pay-per-use model. Client's requests should be processed in an appropriate CP datacenters in a trade-off relation between price and performance. The appropriate choice of a CP datacenters is the responsibility of the cloud-based service broker routing policy which acts as an intermediate between the users and the CP's datacenters. However, due to the distribution nature of the CP's datacenters, these datacenters can be overloaded with the increasing number of users and their requests being served at the same time if the datacenters are unwisely chosen. Therefore, choosing the appropriate datacenter is significant to the overall performance of the cloud computing systems. This paper aims to propose an optimized service broker routing policy based on different parameters that aims to achieve minimum processing time, minimum response time and minimum cost through employing a searching algorithm to search for the optimal solution from a possible solution space. A simulation-based deployment of the proposed algorithm along with a comparison study with other known algorithms form the field, confirms the ability of the proposed algorithm to minimize the load on service provider datacenters with minimum processing time, response time and overall cost. [ABSTRACT FROM AUTHOR]

A swift improvement in the development of technologies in this communication era has created an enormous traffic comprising of multimedia data to cloud networks. The multimedia applications are very sensitive to quality of service (QoS) parameters. The throughput of packets is proportionate to the quality of the received multimedia data. The aim of this paper is to improve the throughput of multimedia data particularly for the smart cloud networks by fragmenting the packets into optimal size. The optimal fragment size for standard encoding rates is calculated using soft computing algorithms and other encoding rates are calculated by regression using least squares method. An improvement in the throughput of packets and decrease in calculation time is demonstrated using experimental results and simulation. [ABSTRACT FROM AUTHOR]

How to dynamically deploy web cluster so as to reduce energy consumption and mean-while satisfy performance requirements is an urgent problem to be resolved. In this paper, we propose an online energy-efficient deployment strategy to minimize cluster's energy consumption on the premise of guaranteeing server's CPU utilization equal to a given target value. It adopts CPU equivalent continuous Dynamic Frequency Scaling to reduce server power. First, we propose an approach of CPU utilization guarantee. Then, we describe cluster's energy-efficient deployment problem as a constrained Mixed Integer Programming problem. Compared with similar works, our variable definition manner can reduce variable number significantly. Finally, we propose an improved differential evolution algorithm to solve the problem. Because of few variable number and high solving efficiency, even if applied to large-scale clusters, our strategy can still dynamically deploy the cluster online. Evaluation results verify the feasibility and effectiveness of the proposed deployment strategy. [ABSTRACT FROM AUTHOR]

The accurate detection of user interest data in cloud computing environment can improve the quality of data management. For the accurate detection of user data, we need the adaptive train for the spatial data clustering process obtaining the data clustering objective function, and then complete the accurate detection of a characteristic data. This paper proposed the method of user interest data detection in cloud computing environment based on spatial autocorrelation and differential evolution theory. The method used spatial autocorrelation theory of neighborhood object to obtain the distance between outlier spatial data and its neighborhood spatial data, clustering all data for obtaining the data mean reference point, fitting the generated data mean reference point. The higher-order cumulant feature of data row was extracted. We used the differential evolution theory for the adaptive training on the clustering process of spatial data, the data clustering objective function was obtained. On this basis, we complete the user interest data detection. Experimental results show that the proposed method can accurately detect the user interest data in data space, and the false alarm rate of the proposed method is well below the traditional method. In the case of the same amount of data, the running time of the proposed method is lower than the traditional method. The proposed method has high detection accuracy and greatly improves the quality of data management in the cloud computing environment. [ABSTRACT FROM AUTHOR]

Route calculation is a problem in real life of every day. Multilevel road network is an effective way to speed up the optimal route calculation on large size road networks by separating the original road network into subnetworks and pre-computing new sections on the higher level network. The efficiency of the route calculation depend on how to separating the original road networks. In order to optimize the structure of multilevel networks, an evolutionary-based automatic clustering method, which can automatically search for a proper number as the number of clusters, is proposed to sperate the road network in this paper. Geographic based crossover and differential evolution based mutation are used to enhance the searching ability of the proposed method. Two objectives are considered during the evolutionary process. A practical evaluation is designed to select a proper solution from the Pareto solution set. The proposed algorithm is systematically studied in various sizes of road networks. Experimental results show that the multilevel networks constructed by the proposed approach are effective and efficient for route calculation. [ABSTRACT FROM AUTHOR]

Metaheuristics are gaining increasing recognition in many research areas, computational systems biology among them. Recent advances in metaheuristics can be helpful in locating the vicinity of the global solution in reasonable computation times, with Differential Evolution (DE) being one of the most popular methods. However, for most realistic applications, DE still requires excessive computation times. With the advent of Cloud Computing effortless access to large number of distributed resources has become more feasible, and new distributed frameworks, like Spark, have been developed to deal with large scale computations on commodity clusters and cloud resources. In this paper we propose a parallel implementation of an enhanced DE using Spark. The proposal drastically reduces the execution time, by means of including a selected local search and exploiting the available distributed resources. The performance of the proposal has been thoroughly assessed using challenging parameter estimation problems from the domain of computational systems biology. Two different platforms have been used for the evaluation, a local cluster and the Microsoft Azure public cloud. Additionally, it has been also compared with other parallel approaches, another cloud-based solution (a MapReduce implementation) and a traditional HPC solution (a MPI implementation) [ABSTRACT FROM AUTHOR]

Proportional-integral-derivative (PID) controllers are the most popular control systems equipped in industries due to their simplicity, effectiveness, and functionality. In this article, an adaptive differential evolution (DE) algorithm is presented to tune controller parameters of PID systems. The proposed algorithm uses shift based parameter control and pseudo population reduction procedures. All algorithmic parameters of DE are adapted and no additional parameter is introduced. A set of three typical control instances is taken to study the performance of the proposed algorithm. Four recently reported DE algorithms are chosen as baselines. Through numerical experiment, it turns out that the proposed algorithm yields better performance than the four baseline DE algorithms. Moreover, the proposed algorithm has a better scalability and reliability than other test algorithms. [ABSTRACT FROM AUTHOR]

This paper proposes a transparent and reliable method between the suppliers and consumers for optimal reactive power pricing. The electric power suppliers compute the optimal reactive power using optimal reactive power dispatch problem by considering nodal voltage stability index ‘I’ as one of the constraints. The computed optimal reactive power of the generator is included in the reactive power pricing. The pricing method to the suppliers based on the opportunity cost method is presented and a detailed analysis using 62 bus Indian utility system has been carried out by considering diverse cases. In this proposed pricing method, the services of the cloud technology have been used to provide transparent pricing based on the demands of the consumers. The power demands at the consumers’ site is calculated without the human involvement using the Internet of Things and the same is uploaded in the cloud. In reactive power pricing, the system operator acts as a mediator between the suppliers and consumers. Based on the demand and availability of power, the system operator provides the cost for the service to the consumer through cloud.This paper proposes a transparent and reliable method between the suppliers and consumers for optimal reactive power pricing. The electric power suppliers compute the optimal reactive power using optimal reactive power dispatch problem by considering nodal voltage stability index ‘I’ as one of the constraints. The computed optimal reactive power of the generator is included in the reactive power pricing. The pricing method to the suppliers based on the opportunity cost method is presented and a detailed analysis using 62 bus Indian utility system has been carried out by considering diverse cases. In this proposed pricing method, the services of the cloud technology have been used to provide transparent pricing based on the demands of the consumers. The power demands at the consumers’ site is calculated without the human involvement using the Internet of Things and the same is uploaded in the cloud. In reactive power pricing, the system operator acts as a mediator between the suppliers and consumers. Based on the demand and availability of power, the system operator provides the cost for the service to the consumer through cloud.This paper proposes a transparent and reliable method between the suppliers and consumers for optimal reactive power pricing. The electric power suppliers compute the optimal reactive power using optimal reactive power dispatch problem by considering nodal voltage stability index ‘I’ as one of the constraints. The computed optimal reactive power of the generator is included in the reactive power pricing. The pricing method to the suppliers based on the opportunity cost method is presented and a detailed analysis using 62 bus Indian utility system has been carried out by considering diverse cases. In this proposed pricing method, the services of the cloud technology have been used to provide transparent pricing based on the demands of the consumers. The power demands at the consumers’ site is calculated without the human involvement using the Internet of Things and the same is uploaded in the cloud. In reactive power pricing, the system operator acts as a mediator between the suppliers and consumers. Based on the demand and availability of power, the system operator provides the cost for the service to the consumer through cloud. [ABSTRACT FROM AUTHOR]

For detection and classification of pulmonary nodules, there are two major issues exists in the existing computer aided diagnosis system. First major problem is automatic threshold to segment lungs and nodules. Threshold selection is a critical preprocessing step for medical images. Gaussian approximation based differential evolution has been used to find out the optimal threshold value for segmentation of lungs. Initially, 1-D histogram of the image is estimated using a blend of Gaussian functions whose parameters are calculated using the differential evolution method. Every Gaussian function estimating the histogram characterizes a pixel class and hence a threshold point. Second major problem is to extract the optimized features for classification of nodules. So, a novel gradient intensity feature descriptor for pulmonary nodule classification has been proposed using the multi-coordinate histogram of gradient and intensity based statistical features descriptor. Ensemble bagging trees has been used intelligently using the concepts of ensemble to classify the nodules. We have used standard dataset titled lung image consortium database for the verification and authentication of our proposed computer aided diagnostic (CAD) system. The proposed CAD system gives better results in comparison with existing CAD systems. The sensitivity of 97.5% is attained with an accuracy of 98.7%. [ABSTRACT FROM AUTHOR]