مجله محاسبات و سامانه های توزیع شده
سال دوم شماره 2 (پیاپی 4، پاییز و زمستان 1398)

The optimal resource utilization is an important concern for the organizations and the service providers. The goal of cloud computing is to provide the unlimited reservoir of resources for responding to all the requests coming from the workloads and applications. The IaaS layer is responsible for providing physical and virtual resources (compute, network and storage resources) for upper layers. Current approaches in cloud infrastructure, use three separate management systems for each of these resources. Separate management of cloud infrastructure resources cannot respond effectively and efficiently to the applications and multimedia services which require guaranteed quality of services. The performance of the cloud infrastructure resources relates to the integration of the resource management. In this paper we first describe two resource integration management systems. Then we describe the goals, the criteria and sub-criteria of resource integration management system. At the end, we compare these two systems based on their feature sets and the mentioned criteria and sub-criteria.

Discovering and analyzing of hidden information in complex networks and especially social networks has attracted the attention of many researchers. Community detection is a very important tool for understanding network information and its structure. Many approaches have been proposed, but proposing new methods with high accuracy and high quality are still one of the most exciting and challenging topics in community detection field. Label Propagation Algorithm (LPA) is one of the most popular algorithms in the field of community detection. The random nature and instability in the results of this algorithm make the LPA unreliable. Determining the core nodes and adopting a specific order of nodes based on their importance, more accurate diffusion of the core node label, and improving the way of selecting the correct label, are the main topics that we will discuss in this paper. In this paper, the proposed method which is combination of label diffusion of core nodes and label propagation method, has been able to improve the accuracy and efficiency of LPA by selecting and diffusing the core nodes label and improving the label update strategy. The results of experiments performed on standard and synthetic datasets and comparing the results show that the proposed algorithm performs better than the LPA and NIB-LPA algorithms.

Housekeeping genes are expressed in all biological conditions and stages of tissue growth. A housekeeping gene is the basic gene required for maintaining the cell's primary function and is expressed in most cells of an organism to produces corresponding proteins. Normally, some genes are not expressed in different conditions and steps of the cell cycle that can determine the dynamics of the presence of a protein and consequently transient or stable interactions. Stable interactions play an important role in the determination of protein complexes. One of the challenges in system biology is to determine the appropriate thresholds to determine active proteins at each time point and to identify stable interactions. Our goal in this paper is to determine the unique threshold for each gene to characterize housekeeping genes and stable interactions. Determination of the specific threshold for each gene is determined by using the firefly optimization algorithm with the attraction function based on the combination of the standard protein complex and co-expression of genes. Experimental results on the existing dataset show that the thresholds created by the proposed method had better results than the previous methods.

Simply put, the Internet of Things (IoT) is defined as the ability to connect all the objects around us, through the internet. The Internet of Things is composed of two words, First internet that shows IoT is based on IP and then objects that represent objects and disparate devices wanting to connect and their data exchange. In this technology, billions of objects can communicate with each other and with their surroundings and to exchange information. The exchange of information between objects and human life has changed quickly. The IoT as an emerging technology by combining digital and physical realms, has expanded access to information technology and improved human life under his influence. The IoT therapy is a technology that incorporates wireless sensors into medical equipment, then equips and connects patients and hospital staff and equipment to the Internet to revolutionize modern medicine. Although this technology, with all its dimensions, is based on its core network and connections, It is a combination of different fields of knowledge in terms of research. In this article, we first examine the various connection and interaction technologies used in the IoT to exchange information between devices, and then inspect the applications, challenges, and implementation of the IoT in the field of medical sciences. Finally, given the growing prevalence of Coronavirus (covid-19), smart medical technologies that have been helpful, to minimize contact with the patient will be discussed.

The growing amount of information on the internet has made it difficult for many web users to make the decision-making and selection of information, data or goods. Recommended systems are designed to address this challenge and try to offer specific and useful information with respect to user tastes and past user experiences. So far, many offering systems have been developed in a variety of applications including movies, music, books, hotels etc. Choosing the right trip, the hotel proposal and so on, with regard to the individual's budget usually have a lot of difficulties and concerns for users and generally takes a lot of time and energy. In this paper, a travel and hotel recommendation system is developed which is constructed from combination of different filtering methods to maximize accuracy. The system is considering the current user's preferences, the quality of the service packages and past experiences of the same users with the current user in order to providing more accurate suggestions. It also eliminates the cold start problem.

In order to provide quick selection and decision making in the field of mobile phones marketing, a knowledge-based mobile recommender system along with a personal interest questionnaire is proposed in this research. Initially, the mobile technical data was collected, cleaned, selected, and converted, then assigned labels to the data according to several extracted rules according to users' needs. Then a hybrid model was proposed using classification algorithms such as decision tree, naive Bayes and generalized linear model. This hybrid model along with deep learning algorithms and random forest were used to classify cell phones separately. Finally, these models were evaluated in terms of accuracy. Users can choose their preferred mobile phone using the suggested model and enter their preferences using the Personal Interests Questionnaire.

Soft computing based on spectral spatial features are introduced by this pater to analyze hyperspectral images. In the proposed pattern recognition algorithm, after pre-processing, three pre-processing steps are applied, including the construction of initial probability maps, filtering of probabilistic maps, and maximizing the probability to the soft computing classification class. In this article, some of the spectral and spatial features of pre-processed remote sensing images are offered to the Mamdani fuzzy inference system. The optimal features for applying to the fuzzy inference system are selected using the genetic algorithm and SVM. The efficiency of the proposed pattern recognition algorithm is assessed using some real datasets and several evaluation criteria. The empirical results efficiently show the performance of the proposed method in classifying hyperspectral and computational intelligence algorithms.

The basis of many phenomena in the various sciences are inherently nonlinear, formed by partial and ordinary differential equations. With the exception of a limited number of these equations that have precise analytical solutions, most of these problems do not have precise solutions, which must be solved by new coding-based methods based on software such as Mipple and MATLAB. In this paper, using a new generalization to shape the answers in the Bacland conversion method, we use Mipple software to express new salutary answers to the generalized nonlinear equation of Radacricians-Cando-Lakshminen.

Modern life has changed the way of life in the world today and has caused many diseases, including overweight and diabetes as a consequence of inactivity.Smart computer systems, including recommendation systems, have found increasing use in a variety of fields, including medicine, and can provide various consulting services as a means of filtering large volumes of information under the supervision of physicians. In this study, we propose a model for a smart system that evaluates the fitness conditions and knowing the nutritional needs of the users and proposes a suitable diet plan according to physical condition such as age, height, weight, etc.

In this research, the weight of the poultry farms and the feed conversion ratio are intelligently calculated and sent to the user by SMS. Weight measurement is done by processing the captured images. The feed conversion factor or FCR is an important parameter in the herd breeding process that is directly related to weight and is obtained with high accuracy by the system by calculating the weight. Images captured by a 50-Mpixel camera connected to the processor are recorded and processed in the Raspberry Pi, which is the processor, by a program written in QT software. Information is sent by the GSM SIM800C module, which communicates with the processor via Arduino. Image processing speed is very fast, and the processing operation calculates the weight by calculating the number of target pixels in the image, which is linearly related to the weight. The error rate of 7.5% indicates the high accuracy of this system.

Data privacy is a major concern in a cloud computing environment that uses the Database-as-a-Service model. Nevertheless, the existing encryption schemes are only partly homomorphic and are designed to enable for one particular method of computation to be carried out on encrypted data. To address these concerns, we have recently proposed new partially homomorphic schemes that preserve data privacy in a cloud database while still enabling encrypted data to be processed. The proposed schemes allow the processing of queries with an order-preserving partial homomorphic encryption scheme. We've shown that our schemes are a realistic framework that protect data privacy with reduced overhead.

In recent years, research into motion mining and tracking of moving objects in real-time have attracted the attention of many researchers. Therefore, a new model for motion mining based on a combination of sequential and parallel tracking strategies has been presented in this paper in order to take advantage of them and reducing their shortcomings simultaneously. In fact, combining tracker-level model helps to choose the right motion mining algorithm based on input data features, and also reduces tracking error by synchronizing tracker activity with parallel and series strategies simultaneously. In comparison with other existing solutions, this model provides important advantages such as decreasing the response time, improving the speed and increasing accuracy for tracking moving objects in the higher layers.