جستجوی مقالات مرتبط با کلیدواژه "genetic algorithm" در نشریات گروه "فناوری اطلاعات"

The possibility of developing intelligent systems compatible with the environment has increased by expanding computer engineering fields. Self-adaptive systems are one of the types of software systems that change their behavior according to system conditions and environmental conditions and adapt themselves to it. Although the evaluation of the performance of these systems has been covered in most researches, the evaluation of their quality has remained closed. Therefore, designing an approach is an essential issue to evaluate the quality of self-adaptive systems. The first challenge, the quality indicators of these systems are not fixed, specific, and definite parameters. For example, one of the qualitative indicators for a self-adaptive system is the ability to run software on different operating systems. This indicator may have different degrees of importance for different experienced individuals. The next challenge, some qualitative indicators are not mathematical variables, but rather a linguistic variable between users and experts, indicating that these qualitative quantities are fuzzy variables and can be completely formulated by fuzzy logic. In this article, a new genetic-fuzzy type 2 approach has been proposed to evaluate the quality of the systems based on their quality indicators. The fuzzy logic type 2 is used to describe qualitative indicators, and a genetic algorithm is utilized to determine the optimal fuzzy weights of the qualitative indicators.

In the proposed method, it has been tried to compare the self-adaptive systems from two dimensions, including the software dimension and their self-adaptive dimension. This is despite the fact that most of the existing research deals with only one dimension. In order to evaluate the proposed method, a traffic control system called InSync, which is an adaptive traffic control system (ATCS) and contains multiple qualitative factors, has been used. The obtained results confirm the effectiveness of the proposed method due to the lack of scenario generation, the ability to extend the method to all software quality parameters, and its simplicity. Also, the method is more comprehensive than other existing evaluation models. In addition, it has been tried to compare the self-adaptive systems from two dimensions, including the software dimension and their self-adaptive dimension. This is despite the fact that most of the existing research deals with only one dimension.

Travel planning is a critical aspect of tourism, presenting significant challenges for visitors exploring unfamiliar cities. The Tourist Trip Design Problem (TTDP) focuses on optimizing routes for tourists interested in visiting multiple Points of Interest (POIs) to enhance the efficiency of daily sightseeing activities in a city. Our study models the TTDP using the Orienteering Problem (OP) while considering user-specific travel constraints such as time limitations and fixed start and end points at particular POIs. Additionally, an innovative approach is introduced that incorporates user interest levels based on personalized visit duration preferences into the travel planning model. By utilizing genetic algorithms, this approach ensures robust search efficiency and facilitates both global and local optimizations. Empirical evaluations on real-world datasets demonstrate that our proposed algorithm outperforms in achieving diverse and optimized travel plans and is confirmed to be effective in scalability concerning different problem sizes and optimization goals.

Software-Defined Networking (SDN) concept involves separating the control plane from the data plane. The network control is determined by an entity called the controller located on the control plane. Determining the optimal number and placement of controllers on the control plane is known as the Controller Placement Problem (CPP). This article addresses the resolution of CPP using Unsupervised Kmeans (U-kmeans) and Horse Herd Optimized Algorithm (HOA). The U-kmeans algorithm clusters switches and determines the number of controllers. Since the CPP problem is discrete, the HOA algorithm uses genetic operators, called Modified HOA (MHOA). The next step of this article involves finding the optimal location for each controller within its cluster using MHOA. To improve the convergence rate, MHOA utilizes an Elite Opposition-based Learning (EOBL) strategy. The effectiveness and scalability of the proposed algorithm are evaluated through simulation tests on various networks. The results show that the proposed method outperforms other state-of-the-art algorithms regarding metrics such as end-to-end delay, load imbalance, and energy consumption. In particular, the proposed method reduces load imbalance by 9.66%, end-to-end delay by 19.65%, and average energy consumption by 8.43%.
 

Currently, organizations use data mining and business intelligence tools and techniques to analyze the behavior of their customers. Customer segmentation is a broad analytical tool used to identify distinct groups of customers. Because static segmentation leads to missing important patterns and trends in customer behavior over time. In this research, a model is presented that displays the behavior of each customer as a time sequence of the variables of purchase novelty, number of purchases, purchase amount, and customer cost. In fact, it also considers the time dimension of customer behavior. Then, using the genetic algorithm, optimal weights are found for each feature, and customers are segmented with clustering algorithms. The data used in this research is related to the transaction data of a payment service company during a period of thirty months. The results indicate that the best clustering result is achieved using spectral clustering algorithm by computing silhouette and Calinski-Harabasz metrics. These findings demonstrate that with optimal weighting, the genetic algorithm has been able to combine the features in a way that improves the silhouette metric to 0.91.

With the high potential that cloud computing has for remote data storage and processing, Cloud computing has introduced a new computing technology for storing and processing data. In cloud service environments, virtual machines from multiple organizations are placed on a physical server, which increase the efficiency of virtualization. In such a scalability infrastructure, we have some bottlenecks and congestion in our cloud without managing traffic. So, we need the solutions to balance the load and distribute tasks among other processing servers. We present an advanced solution based on optimized genetic algorithm for scheduling and balancing load in cloud infrastructure in this paper. We add some parameters to the algorithm to check resource status before scheduling. In fact, the proposed technique prevents overload or underload on the servers by optimally assigning tasks to processing servers and the tasks of high-load and congested servers are transferred to another server using live migration of the virtual machine, in order to increase the load balance of the cloud infrastructure At the end, the proposed solution is evaluated through the Cloudsim simulator and by testing the volume of more than a thousand virtual machines on the Plant Lab data. The results of the simulation results show that the proposed solution has been able to violate the service level agreement compared to the AMUT  and EQVS methods on average 46%, the energy consumption criterion on average by 18 % and averages number of virtual machine migrations improved by 24%.

A critical issue in designing Wireless Sensor Networks (WSNs) is to prolong their lifetime considering the limited energy resources of the nodes. One approach to mitigate this problem is to apply Dynamic Modulation Scaling (DMS) technique. By adjusting the modulation level of the nodes, the amount of energy exhausted and the required time for data transmission are balanced. Accordingly, the energy consumption throughout the network reduces substantially. The routing problem should be considered along with modulation level adjustment to enhance performance. The existing DMS-based algorithms either did not concern the routing problem or proposed greedy and low-quality solutions. Hence, we propose an effective algorithm which comprises two phases, namely routing and determining the modulation level of nodes. The proposed scheme exploits genetic algorithm to improve performance. Simulation results demonstrate that the proposed algorithm improves the average exhausted energy by sensors, the standard deviation of consumed energy by sensors, and the network lifetime, by 72%, 88%, and 112%, respectively.

Fog computing has emerged as a promising technique to provide agile and pervasive computing services to the Internet of Things devices (IDs) and to support complicated IoT applications. Fog computing brings computation resources to the edge of the network, near to the IDs, and provides low-latency services to users. By offloading computational tasks to fog nodes having greater computing capacities, can address the contradiction between the limited battery capacity of IDs and high computational intensity demand of tasks. Hence, the quality of service (QoS) demands of users can be fulfilled. Although task offloading to fog nodes leads to saving in energy consumption in the battery of IDs, it causes to increase in task completion time due to occurred delay in transmitting the task to the edge of the network. In this paper, to balancing the trade-off between energy consumption and task completion time, a task offloading scheme is proposed. The main objective of the proposed scheme is to minimize offloading overhead in terms of the weighted sum of energy consumption and task completion time by optimizing offloading decision, the destination of offloading, and computation resource allocation. We employ fuzzy logic to determine the weighting coefficient effectively. Task offloading to fog nodes is formulated as a mixed-integer nonlinear programming (MINLP) problem, which is NP-hard. A sub-optimal algorithm based on genetic algorithm (GA) is proposed to solve the formulated problem. Extensive simulations prove the convergence of the proposed algorithm and its superior performance in comparison with some baseline schemes.

In this paper, a profit-based distribution network expansion planning model will be examined. In this model, some capabilities of the automation system are considered to achieve a plan that is more compatible with the strategic plan to achieve a smart grid. In addition, a method has been used to assess the reliability of a distribution network so that development planning studies can be conducted more efficiently. The objective function of the planning problem is the net worth of the firm. Using genetic algorithm, this nonlinear programming problem of complex integer is solved. The method has been evaluated in a visual experimental network and the results are presented and discussed. The results show that the development plan with automation is more profitable than the conventional development plan.

in visual secret sharing, the secret image is transformed into several share images and distributed among different people. The share images do not contain any information about the original image, and these images are similar to the noise image. When needed, the original image can be retrieved in the presence of all stakeholders and by stacking the share images. In this regard, the appearance of noisy images may attract the attention of the attackers. To solve this problem, meaningful visual secret sharing was presented. In this article, a method for sharing the image with meaningful shares is introduced. There are a number of hyper parameters in the proposed algorithm. To improve performance, an attempt has been made to determine these hyper parameters using a genetic algorithm. The cost function of the genetic algorithm is the difference between the two values of the correct number of bits of the recovered image and the original image and the correct number of integer bits between the cover image and the shares. The proposed method was evaluated using PSNR, MSE, BCR criteria and presented good results on different images with different number of shares.

Nowadays, online review of user’s sentiments and opinions on the Internet is an important part of the process of people deciding whether to choose a product or use the services provided. Despite the Internet platform and easy access to blogs related to opinions in the field of tourism and hotel industry, there are huge and rich sources of ideas in the form of text that people can use text mining methods to discover the opinions of. Due to the importance of user's sentiments and opinions in the industry, especially in the tourism and hotel industry, the topics of opinion research and analysis of emotions and exploration of texts written by users have been considered by those in charge. In this research, a new and combined method based on a common approach in sentiment analysis, the use of words to produce characteristics for classifying reviews is presented. Thus, the development of two methods of vocabulary construction, one using statistical methods and the other using genetic algorithm is presented. The above words are combined with the Vocabulary of public feeling and standard Liu Bing classification of prominent words to increase the accuracy of classification

Shape detection in digital images is one of the most effective subjects in image processing. This paper introduces a method for circle detection in digital images using genetic algorithm. Circle is expressed by a quadratic relation in coordinates screen. For circle detection, finding coefficients of this quadratic relation is the challenge. In the proposed method, three random points are selected on the edges of image. Because of each three points that do not locate on a straight line express a circle in coordinates screen, the quadratic relation coefficients of the circle are considered as the chromosome of genetic Algorithm. After finding the coefficients and drawing the circle, fitness function is calculated by computing amount of overlapping this circle with the edge of image. Then, the polynomial coefficients of new generations are generated by using crossover and mutation operators. Genetic algorithm continues until reaching the final conditions. Results of experiments on some of the images show that the proposed method can find circles on images. Increasing of the success rate in circle detection on image and exact detection of center and radius of circle are the contribution of this article in comparison with studied methods.

The LSB matching method or LSBM is one of the simplest methods of steganography that has been proposed relatively successful attacks for its discovery. Visual image quality (imperceptibility) and lack of discovery by steganalysis attacks are two important criteria for any method of steganography. The main purpose of this paper is to provide a LSBM-based approach that is superior to LSBM in these two criteria. In the proposed method, the cover image is blocked and selected the best embedding sequence for each block using the genetic algorithm and the Linear Congruential Generator (LCG). The best sequence contains pixels whose LSB correspond most to data bits. The second step is to use the LSBM in the pixels of this embedding sequence. If the secret bit does not match the pixel’s LSB, the pixel value should be incremented or decremented randomly by one unit. To make these random selections, a genetic algorithm has been used, so that the block has the least change in histogram compared to the original block. Comparing the parameter of visual image quality and the accuracy of the attacks in discovering this method, indicates the proper improvement of these criteria compared to the LSBM method.