نشریه محاسبات نرم
سال دهم شماره 1 (پیاپی 19، بهار و تابستان 1400)

Today, the many benefits of cloud computing have led many businesses, large and small, to use cloud services to reduce their costs. In the meantime, there are some barriers to using cloud services, one of the biggest of which is security attacks affected by the supervisor. When a direct attack on a user is carried out on a supervisor, it may also indirectly attack other users' virtual machines. Meanwhile, the conflicting goals and interests of cloud service users and attackers make it difficult for cloud service providers to decide to invest in cloud server security modules. Therefore, in this article, using game theory, a suitable solution for deciding to invest in one of the security modules for each of the actors is presented. Also, using the iterative game model, all Nash equilibria have been extracted and analyzed. The results show that game theory can be well used in making the right decision and finding the right balance to make decisions about investing in the field of applied security. Based on the simulation results, it can be said that in repetitive games with a probability of repeating the game between 0.2 and 0.8, predetermined investment strategies or non-investment can lead to an appropriate Nash equilibrium and Pursue maximum benefits for cloud service users.

Since the formation and Direct solving the governing equations requires high time and computational cost, this study seeks to provide an equation free model based on deep learning algorithm that simulates steady state heat transfer in two-dimensional space and a relatively large size using order reduction method. Principal component analysis is a linear method and autoencoder is a nonlinear methods. The results of comparing their performance on different data sets showed that in reducion of order to very low dimensions, autoencoder and in reducion of order to very high dimensions, principal component analysis has a higher accuracy. Of course, the number of dimensions to order reduction and the characteristics of the data set such as size and number of dimensions of the data will affect the accuracy of the dimensional reduction. These two methods were used to order reduction of thermal data in order to faster simulate the phenomenon of Steady State Heat Transfer and were compared with a model based on convolutional neural network with a number of layers and multiple filters. The results showed that the models based on order reduction methods have much less computational volume and simulation time, and the outputs obtained from them, especially the model based on the autoencoder method, have a much higher accuracy.

The most critical challenge in edge computing is to increase the number of users of the service provided without changing the complexity of the problem and imposing further delays for servicing the cloud-edge network. First, every user will request the service from their edge server, which edges provide it. If there is no service in the edge cloud, the inquiry is made from the adjacent edge server based on the distance. The nearest edge clouds that can connect to it wirelessly provide the desired service. If it fails to receive the service from them, the service request enters the backhaul link and searches for the requested service on other edge servers. Finally, if it still fails to receive the service, it will receive the service from the cloud. Due to the complexity examined, the strategy presented in the paper provides the conditions to increase the number of service users without changing the service delay. The average improvement for users served to all users in the proposed algorithm. However, 0.4% relative to the greedy and 1.1% relative to the maximum flow, the proposed method shows a 20% and 22% improvement over the maximum flow and greedy algorithm in users' scheduling, respectively.

Breast cancer is currently the most frequently diagnosed cancer and leading cause of cancer death in women worldwide. Researchers have been researching the best treatment for breast cancer. Their focus has been on preventing recurrence after the initial treatment of patients. Choosing the appropriate model for survival time analysis is the main challenge in the survival analysis of these patients. In this applied research, using graph-based semi-supervised learning method a new survival analysis model is proposed for analyzing the survival of breast cancer patients. Also, a dataset of 3833 patients with breast cancer who were followed up for 5 years was used. The clinical and pharmacogenomics information, including the results of Tamoxifen use and effect on the treatment of invasive cancer, was also available and used. To validate the proposed model, by introducing the evaluation parameters and simulation of the models in MATLAB software, the model performance is compared with previous models of survival analysis. The results demonstrate that by applying the proposed model in predicting invasive disease-free survival time and when a combination of the clinical and the pharmacogenomics features were used the estimation accuracy was 14% higher than when only the clinical features were used. Moreover, the estimated accuracy was 15% higher than when only the pharmacogenomics features were used. The proposed survival analysis model has a high capability for identifying survival risk and high accuracy in predicting patients' survival time.

medical intelligence detection systems have been changed and also faced with some challenges. Breast cancer diagnosis and classification is one of these medical intelligence system. Early detection can increases the options for curative treatment. There are a variety of screening techniques available to detect breast cancer such as mammography, magnetic resonance imaging and ultrasound. These images have been presented for a variety of applications and different processing techniques to date based on the type of cancer diagnosis. This research used MIAS mammography image dataset and try to diagnose and classify benign, malignant and suspicious masses based on image processing and machine learning techniques. So, a new developed approach proposed which at first, apply pre-processing for noise reduction and image enhancement based on Quantum Inverse MFT, and then image segmentation with Social Spider Algorithm based on two features such as brightness and edges apply. Then two main parts of diagnosis and classification apply which based on ELM and MPM-ELM. Obtained results presented that proposed approach have better performance in comparison to others based on some evaluation criteria such as accuracy, sensitivity, specificity and also ROC and AUC

Abstract- In recent studies, there has been a lot of research on cloud federation due to the increase in profits of cloud providers and resource utilization. Each of the studies has helped develop this area by improving one or more features of the cloud federation. However, these studies suffer from a lack of attention to integrity issues. Integrity means maintaining a motivating platform for cooperation and member participation in a way that no member is more profitable in exchange for deviating from the truth. This paper has designed and implemented a review system to review and analyze related studies from the integrity perspective in the cloud federation. The studies that explicitly or implicitly support integrity have been selected and discussed in the strengths and weaknesses. Obstacles and challenges in providing integrity and suggestions for improving it in the cloud federation have also been discussed. According to our knowledge, this paper is the first Systematic Literature Review (SLR) on the recent cloud federation studies from the integrity perspective.

Diabetic retinopathy is one of the most common causes of blindness in the world. Identifying blood vessels from retinal images plays an important role in diagnosing eye diseases such as diabetic retinopathy. Extraction of retinal blood vessels is done manually by a doctor, which is time-consuming and difficult and is associated with error due to individual dependence. Detecting small veins in retinal images is difficult due to low contrast and rotation. In this article, a new method for extracting blood vessels from digital retinal images is presented. This method consists of three basic parts. The first part deals with the removal of noise in retinal images. The second part deals with the extraction of the central lines of the vessel, and finally, using the method of expanding the area and removing the noise, blood vessels are extracted in the retinal images. The proposed algorithm is applied to the images of the DRIVE test set and the implementation results show that this algorithm has achieved an accuracy of 0.96578% in this data set.

Social networks have found a significant place in society. The large number of users of these networks and the high volume of data that is exchanged through these networks daily have made these networks a good opportunity for advertising companies. Due to the budget constraints, companies are forced to select only a limited number of users to begin the propagation process. In order to be successful in advertising and informing the majority of the community, the selected collection must be selected among influential people. Many studies have addressed the issue of selecting the optimal subset of influential people. But in the majority of them, it is assumed that the influence between people is positive and the negative influence of people on each other is not considered. Therefore, in this study, a method proposed for selecting the optimal subset of influential people according to the positive and negative influences between individuals. In the proposed method, first for each member of the community, a set of positive and negative influences of one and two steps is obtained. Then, using multi-criteria decision making method, a method for selecting the best people is presented. The Wikipedia data set was used for the evaluations. The results obtained for the spreading power of the set selected by the proposed method and other methods indicate that although in some cases some methods have more positive propagation than the proposed method, but the proposed method in all Scenarios has a much better performance in terms of negative emission.