Search

Multi‐object tracking (MOT) can be considered as an interesting field in computer vision research. Its application can be found in video motion analysis, smart interfaces, and visual surveillance. It is a challenging issue due to difficulties made by a variable number of objects and interaction between them. In this work, a new method for online MOT based on time and frequency domain features is presented. The features are obtained from the wavelet transform and fractal dimension. The modified cuckoo optimization algorithm is utilized for feature selection, which has the ability such as fast convergence and global optima finding. The features are given for learning vector quantization, which is a supervised artificial neural network (ANN). It is used to classify the dataset. To evaluate the performance of the presented technique, simulations are performed using the ETH Mobile Platform and VS‐PETS 2009 datasets. The simulation results show the superiority of the presented technique for MOT compared to earlier studies in terms of accuracy. The mostly tracked values for the datasets are 74.3% and 97.2%, which leads to at least 4.2% and 2.5% better performance according to the other methods, respectively.

In a cloud computing environment, there are many providers offering various services of different quality attributes. Selecting a cloud service that meets user requirements from such a large number of cloud services is a complex and time-consuming process. At the same time, user requirements are sometimes described as uncertain (sets or intervals), something which should be taken into account while selecting cloud services. This paper proposes an efficient method for ranking cloud services while accounting for uncertain user requirements. For this purpose, a requirement interval is defined to fulfill uncertain user requirements. Since there are a large number of cloud services, the services falling outside the requirement interval are filtered out. Finally, the analytic hierarchy process is employed for ranking. The results evaluate the proposed method in terms of optimality of ranking, scalability, and sensitivity analyses. According to the test results, the proposed method outperforms the previous methods.

Due to the threat and increasing trend to diabetes, different approaches to diagnose it have been proposed, so that classification is one of the main techniques. In this article ultimate aim is designing a novel system to diagnose diabetes. To this end, we use an ensemble classifier to apply support vector machine (SVM), k-nearest neighbor (KNN), and whale optimization algorithm (WOA). WOA is responsible for generating weights for each classifier to improve the accuracy of the diabetes classification. For our empirical study, we gathered a dataset of diabetes from medical centers in Iran. The implementation results showed that the designed ensemble classifier achieved the accuracy rate of 83%, which means it improved the accuracy of the best preceding classifier about 5%. Moreover, the designed ensemble classifier based on WOA improved the accuracy about 1% in comparison with PSO that is preceding the WOA in terms of accuracy level.

Component-Based Software Engineering (CBSE) is an approach to building and developing software systems based on software components. In component-based software systems, there are various software components, including Commercial off the Shelf (COTS) and in-house components. Software developers can build their desired software component as in-house or COTS. The problem of deciding optimally between COTS and in-house components is one of the major challenges of software developers, which is known as the component selection problem. This can be resolved by evaluating the criteria for optimality in component selection and then solving the component selection problem by optimization techniques. In this paper, an attempt was made to optimize the component selection problem through the multi-objective optimization by maximizing the Fuzzy-Intra Coupling Density (Fuzzy-ICD) and functionality as objective functions, and also taking into account budget, delivery time, reliability, and Fuzzy-ICD as constraints of multi-objective problems. Fuzzy ICD is a more accurate criterion to calculate the relationship between Cohesion and Coupling of components, which is obtained through the fuzzy computing of each of them, based on the Meyers classification. Thus, after a two-criterion optimization model formulation, this optimization problem was solved by fuzzy multi objectives approach. Finally, the proposed method was evaluated by performing the case study of financial-accounting system. Comparison of the results showed that the proposed method could select optimal components with maximum functionality and Fuzzy-ICD and fewer rates of time and Budget (0.29, 0.43, 1.1 s, and 88$ were the improved rates of functionality, Fuzzy-ICD, time, and budget, respectively).

In recent years, the use of directional sensor networks (DSNs) has continued to rise increasingly, which is due to their extensive use in many situations. In such networks, the main problem is how to cover the targets distributed in a defined area and simultaneously prolong the network lifetime as much as possible. The reason of this problem is the limitation of both sensing angle and energy resource of sensors in such networks. This problem gets more complex in cases where targets need to be covered by more than one sensor direction (i.e., each target needs to be monitored for at least k times). This problem is generally known as target k -coverage problem which its NP-completeness has been already proved in the literature. The k -coverage problem can be considered in over-provisioned and under-provisioned environments. In both of these environments, especially in the latter, it is important to create a balanced coverage, as these environments do not have enough sensors to monitor all targets for k times. Thus, in this paper, we proposed a genetic-based algorithm to solve the problem in over-provisioned environments, then developed the proposed algorithm in another way to solve the problem in under-provisioned networks so that it uses the minimum number of sensors. many experiments were performed to test the efficiency of the proposed algorithm, and the obtained results showed the high capacity of the proposed algorithm in solving the k -coverage problem in both environments.

One of the major problems in microarray datasets is the large number of features, which causes the issue of "the curse of dimensionality" when machine learning is applied to these datasets. Feature selection refers to the process of finding optimal feature set by removing irrelevant and redundant features. It has a significant role in pattern recognition, classification, and machine learning. In this study, a new and efficient hybrid feature selection method, called Ga

One of the important issues under discussion connected with traffic on the roads is improving transportation. In this regard, spatial information, including the shortest path, is of particular importance due to the reduction of economic and environmental costs. Here, the constrained shortest path (CSP) problem which has an important application in location-based online services is considered. The aim of this problem is to find a path with the lowest cost where the traversal time of the path does not exceed from a predetermined time bound. Since precise prediction of cost and time of the paths is not possible due to traffic and weather conditions, this paper discusses the CSP problems with fuzzy cost and fuzzy time. After formulating the CSP problem an efficient algorithm for finding the constrained optimal path is designed. The application of the proposed model is presented on a location-based online service called Snap.

Regardless of the performance of gravitational search algorithm (GSA), it is nearly incapable of avoiding local optima in high-dimension problems. To improve the accuracy of GSA, it is necessary to...

In incremental software development approaches, the product is developed in various releases. In each release, a set of requirements is proposed for the development. Usually, due to lack of funds, lack of time and dependency between requirements, there is no possibility to develop all the required requirements. There are two conflicting objectives for choosing an optimal subset of the requirements: increasing customer satisfaction and reducing development costs. This problem is known as the next release problem (NRP) and is categorized as an NP-hard problem. Unlike the standard version of the NRP, we formulate this problem as a restricted multi-objective optimization problem. There exist metaheuristic algorithms for solving this problem performed as serials. In this paper, we introduce a parallel algorithm based on the master–slave model in order to improve the quality of the solutions. Based on the criteria of multi-objective problems, the quality of the obtained solution is compared with several metaheuristic algorithms. Two scenarios and two different datasets are used for experiments. Results indicate that the proposed method in the first scenario would highly improve the quality of solutions. Moreover, the method reduces execution time significantly through improvement in the quality of the solution in the second scenario.

A number of software programms are developed in several releases. Before developing any new release, a set of requirements is suggested for inclusion in the release. Having multiple constraints, it...

The memory management in the operating system includes a part called the page replacement algorithms. Replacement algorithms in environments that require high-performance computing are considered as an important issue. For example, these algorithms are very important in cache management in microprocessors, web caching, replication strategies in distributed information systems and so on. Due to the important role of replacement algorithms in overcoming the problem of performance caused by the difference in processor speeds and memory, many algorithms were proposed. Most of them are the developed schemes of the least frequently used (LFU) and least recently used (LRU). Although most proposed designs can solve the LRU and LFU defects, they are implemented in a difficult way. The most important advantage of LRU and LFU is their simple implementation. This research proposes a page replacement algorithm that is simple to implement. The algorithm, which uses three parameters to cluster cache pages, is called the fuzzy page replacement algorithm. Whenever a miss occurs, it selects a page of the cluster with the lowest priority which has the smallest Euclidean distance with its center and then exits the cache. The most significant advantage of the algorithm is using the FCM (fuzzy c-means) algorithm to cluster pages, resulting in better replacement and hence higher memory performance.

An uncertain Business Process Management System (BPMS) capability is Business Processes (BPs) management in the presence of uncertain factors. This ability should be defined by different uncertain computer-based components inside the classic BPMSs operations. This study proposed autonomous and combinatorial optimal process management architecture to increase the ability, flexibility, and accuracy of uncertain processes management. The autonomous architecture based on the bi-level optimization approach has been constructed inward a meta-model of multi-agent system technology, optimal Neural Network and Cellular Learning Automata in different agents. A case study of an uncertain business process evolving the closed loop supply chain was studied. The results of the simulated case and the statistical evaluation of it, have been demonstrated the robustness and accuracy of this new proposed architecture.

In role-based access control (RBAC), users gain access to predetermined roles and permissions. Thus, desired results are not achieved in emergency situations through policy in RBAC. In emergency situations, users should sometimes gain access to resources not authorized in normal situations. To increase the flexibility of access control, Break the Glass (BTG) policy was proposed. It allows users to break or override access controls, while every operation is documented to create maximum responsibility for users. Users with BTG access have maximum freedom to override the access controls and constraints of the model. In this paper, the flexibility of RBAC is enhanced by proposing an Emergency RBAC (E-RBAC), which uses BTG policy for managing the system in emergency situation. However, separation of duty (SOD) constraint is included to control and limit user access in this situation. Then, an administrative model is proposed to manage large E-RBAC systems. An administrative model reduces excessive burden for an administrator in large E-RBAC systems. At the next stage, E-RBAC is illustrated with medical and drug-dispensation scenarios and is then implemented through Alloy (the first logic language) so as to analyze the validity of model specifications.

In recent years, the directional sensor networks have been attractive to researchers due to their wide and different applications. These networks normally contain a number of self-configurable directional sensors holding adjustable spherical sectors with limited angle. One of the most significant problems in such networks is how to monitor the targets scattered in these networks using sensors with adjustable sensing range and, at the same time, maximize the network lifetime. This problem is recognized as Maximum Network Lifetime With Adjustable Ranges; it has been already proved as an NP-complete problem. As an efficient solution to this problem, the present paper proposes a target-oriented GA-based algorithm that can form cover sets comprising sensors with appropriate directions and sensing ranges in a way to desirably monitor all targets in the network. We examined the efficiency of the proposed algorithm by comparing its obtained results with those of a greedy-based one introduced recently in literature. The comparative results confirmed the efficient performance of the proposed algorithm and also its superiority over the greedy-based algorithm in terms of extending the network lifetime.

Energy consumption has been one of the main concerns to support the rapid growth of cloud data centers, as it not only increases the cost of electricity to service providers but also plays an important role in increasing greenhouse gas emissions and thus environmental pollution, and has a negative impact on system reliability and availability. As a result, energy consumption and efficiency metrics have become a vital issue for parallel scheduling applications based on tasks performed at cloud data centers. In this paper, we present a time and energy-aware two-phase scheduling algorithm called best heuristic scheduling (BHS) for directed acyclic graph (DAG) scheduling on cloud data center processors. In the first phase, the algorithm allocates resources to tasks by sorting, based on four heuristic methods and a grasshopper algorithm. It then selects the most appropriate method to perform each task, based on the importance factor determined by the end-user or service provider to achieve a solution designed at the right time. In the second phase, BHS minimizes the makespan and energy consumption according to the importance factor determined by the end-user or service provider and taking into account the start time, setup time, end time, and energy profile of virtual machines. Finally, a test dataset is developed to evaluate the proposed BHS algorithm compared to the multiheuristic resource allocation algorithm (MHRA). The results show that the proposed algorithm facilitates 19.71% more energy storage than the MHRA algorithm. Furthermore, the makespan is reduced by 56.12% in heterogeneous environments.

In agile methods, software products are developed in several releases. In each release, a new set of requirements for development is proposed. Due to technical and non-technical problems, it is almost impossible to develop all the proposed requirements in the next release. The select of an optimal subset from among all the proposed requirements has become an important problem for the developer team. The aim of this problem is to select an optimal subset from among the requirements for product development in the next release so that it has the highest satisfaction to the clients and the lowest cost for the manufacturing company. Since this problem faces two conflicting objectives and several constraints, it is placed in the NP-hard problems category. In this paper, we intend to formulate this problem for the first time as a fuzzy multi-objective optimization problem. We intend to use an artificial chemical reaction optimization algorithm to solve this problem. In the implementation stage, we make use of five interactions between requirements as one of the constraints of the problem for the first time. Two randomized fuzzy synthetic datasets are used to do the experiments. The results of the proposed algorithm are evaluated using three criteria of multi-objective problems. The results and diagrams of the proposed algorithm are very reliable and can help the developer team to make a decision.

Data are usually stored in databases with temporal components and quantitative values. In temporal databases, each data item has its own exhibition periods. In other word, early items have a longer lifespan. Item lifespan must be considered in mining process for fair and accurate results. Otherwise, extracted patterns may not exhibit the associations of items correctly enough. Some efficient algorithms have been thus proposed for extracting fuzzy association rules from quantitative temporal transactions. However, the membership functions affect rule discovery. Learning or tuning the membership functions for fuzzy temporal association rules mining is thus necessary. In this paper, we combined fuzzy temporal mining concepts into a fuzzy-evolutionary approach and designed an efficient fuzzy temporal-evolutionary mining process based on bees algorithm to find suitable membership functions for fuzzy temporal mining problems before searching for temporal frequent itemsets and fuzzy associations. Experimental results show that the proposed method exhibits good performance with respect to the effectiveness of the obtained solution.

Feature selection refers to selecting optimal feature subset with effective data preprocessing policy in making high dimensional data for diverse pattern recognition problems. The aims of feature selection are enhancing accuracy, improving the evaluation performance, and finding the smallest effective feature subset. In this study, ensemble feature selection method is adopted based on an assumption indicating that a combination of several feature selection methods obtains more robust results than any individual feature selection method. Accordingly, when carrying out ensemble feature selection, a combinational method should be used to combine rankings of features from diverse algorithms into an individual rank for each feature. It is also required to set a threshold to acquire a functional subset of features. In this work, a three-step ensemble feature selection technique called Automatic Thresholding Feature Selection (ATFS) is proposed. The first step involves diversity generation where multiple rankers are applied to each dataset to generate different feature rankings. Second, output rankings of individual selectors are combined using fast non-dominated sorting that is a combinational method empowering the proposed ensemble with automatic thresholding capability. Third, feature sets are generated to obtain the optimal feature set. Additionally, a new filter method called Sorted Label Interference (SLI) is proposed based on interference between class labels. Both SLI and ATFS are applicable to binary datasets. The performance of SLI and ATFS is at least comparable and often better than the performance of individual rankers and existing ensemble methods. The obtained results also show that the use of ATFS-generated threshold improves not only the performance of ATFS and SLI, but also the performance of other filters and combinational methods.

In recent years, wireless sensor networks (WSNs) have attracted many researchers due to their widely usage in a wide range of applications. One of the most important problems in these networks is energy consumption that has a direct effect on network lifetime. Clustering is one of the most important solutions in order to overcome the problem. Energy resource limitation is a fundamental problem in WSNs and clustering protocols provide suitable procedures in order to enhance network lifetime. However, they impose high energy consumption on cluster heads (CH), and therefore, in each round, the protocol should reform clusters and change CH in order to enhance network lifetime. Although these protocols are proper for clustering, do not guarantee suitable CH selection. In this paper, a novel energy-efficient method is proposed using fuzzy logic and three parameters including the amount of energy in CH, distance from CH to base station, and the number of connections in CH. In fact, we focus on the cluster formation process. The proposed model is compared to the well-known low-energy adaptive clustering hierarchy protocol. Simulation results demonstrate that the proposed protocol improves network lifetime.

A potential mathematical framework for machine learning is multi-linear algebra of the higher-order tensor that can reveal the relationships among multiple factors underlying the observations. Prioritizing the requirements of a project facilitates the process of requirements engineering and involves multifactors. Due to existing time constraints and budget related to projects, by prioritizing the requirements in an appropriate order we can select and apply them more accurately and this causes to increase the quality of software and customers’ satisfaction. In order to prioritize the requirements, there are many approaches that consider different parameters and different view point in their prioritization process. But as far as we know none of them considers the simultaneous effect among entities, namely functional requirements, non-functional requirements and stakeholders in their prioritization process. In this paper, we decided to consider the simultaneous effect among functional, non-functional requirements and stakeholders that have different preferences on requirements by modeling a three-order tensor. Then by applying multi-way analysis, we will obtain appropriate ordered lists of requirements. To evaluate our approach, a controlled experiment has been provided that compares the proposed approach with the state-of-the-art-based approach, analytic hierarchy process (AHP). The results show that our proposed approach outperforms AHP in terms of actual time consumption and ease of use while preserving the quality of the results obtained by our proposed approach.

High security of cloud computing is one of the most challenges to be addressed before the novel pas-as-you-go business paradigm is widely applied over the internet. Trust brings a novel means to improve the security and enable interoperability of current heterogeneous independent cloud platforms. However, there is no special trust evaluation model for cloud computing environment. Hence, this paper presents a new trust model based on fuzzy mathematics in cloud computing environment according to success and failure interaction between cloud entities based on the properties and semantics of trust. To compute trust in cloud systems, an algorithm based on proposed model is given. Simulation results show that the proposed model has some identification and containment capability in synergies cheating, promotes interaction between entities, and improves the performance of the entire cloud environment.

Scheduling algorithms have an essential role in computational grids for managing jobs, and assigning them to appropriate resources. An efficient task scheduling algorithm can achieve minimum execution time and maximum resource utilization by providing the load balance between resources in the grid. The superiority of genetic algorithm in the scheduling of tasks has been proven in the literature. In this paper, we improve the famous multi-objective genetic algorithm known as NSGA-II using fuzzy operators to improve quality and performance of task scheduling in the market-based grid environment. Load balancing, Makespan and Price are three important objectives for multi-objective optimization in the task scheduling problem in the grid. Grid users do not attend load balancing in making decision, so it is desirable that all solutions have good load balancing. Thus to decrease computation and ease decision making through the users, we should consider and improve the load balancing problem in the task scheduling indirectly using the fuzzy system without implementing the third objective function. We have used fuzzy operators for this purpose and more quality and variety in Pareto-optimal solutions. Three functions are defined to generate inputs for fuzzy systems. Variance of costs, variance of frequency of involved resources in scheduling and variance of genes values are used to determine probabilities of crossover and mutation intelligently. Variance of frequency of involved resources with cooperation of Makespan objective satisfies load balancing objective indirectly. Variance of genes values and variance of costs are used in the mutation fuzzy system to improve diversity and quality of Pareto optimal front. Our method conducts the algorithm towards best and most appropriate solutions with load balancing in less iteration. The obtained results have proved that our innovative algorithm converges to Pareto-optimal solutions faster and with more quality.

Use case size point (USP) method has been proposed to estimate object oriented software development effort in early phase of software project and used in a lot of software organizations. Intuitively, USP is measured by counting the number of actors, preconditions, post conditions, scenarios included in use case models. In this paper have presented a Adaptive fuzzy Neural Network model to estimate the effort of object oriented software using Use Case size Point approach. In our proposed system adaptive neural network fuzzy use case size point has less error and system worked more accurate and appropriative than prior methods.

 Abstract— Complexities of object-oriented software such as inheritance and polymorphism make behavior analysis significantly difficult, because the states of the objects may cause faults that cannot be easily revealed with traditional techniques. In this paper, we propose a new approach to object oriented software simulation by mapping the specification written in UML to Colored Petri Net (CPN). By introducing an algorithm to convert UML statechart to CPN, The model can be built in an early phase of the software development process, thus creating the potential for early analysis. Our proposed method considers net-explosion problem and the generated Net covers all instances of objects from different classes in the same hierarchy. A case study is presented to show the benefit of our approach and resulting Net is implemented in CPN-Tools.

Reliability is an important software quality parameter. In this research for computation of software reliability, component reliability model based on SPN would be proposed. An isomorphic markov chain is obtained from component SPN model. A quantitative reliability prediction method is proposed. The component reliability value is calculated according to the transition cumulative probability distribution of markov chain, obtained from the software SPN model. By means of reliability prediction of the whole software, we'll introduce CRMPN. In CRMPN states are component reliability model and transition are marked with components reliability. With this research more complex software could be simplified and reliability of the software could be evaluated effectively. An example is provided for demonstrating the feasibility and applicability of our method. Keywords: Reliability, SPN, Markov Chain, Component based-software

in object-oriented software testing, a class is considered to be a basic unit of testing. Attributes of object-oriented software such as inheritance and polymorphism make behavior analysis and test significantly complicated because the state of the objects may cause faults that cannot be easily revealed with traditional testing techniques. In this paper, we propose a new technique for generating the test case by Colored Petri Nets (CPN), which is an extended version of Petri Nets and usually used to system modeling and simulation. Our method considers net-explosion problem and also our generated Net covers all Instances of Objects from Different Classes in the same hierarchy by introducing new algorithm to convert UML Statechart to CPN. A case study is presented to show the benefit of our approach and resulting Net is implemented in CPN-Tools.