نشریه فناوری اطلاعات و ارتباطات ایران
سال چهاردهم شماره 53 (پاییز و زمستان 1401)

In a business environment where there is fierce competition between companies, accurate demand forecasting is vital. If we collect customer demand data at discrete points in time, we obtain a demand time series. As a result, the demand forecasting problem can be formulated as a time series forecasting task. In the context of time series forecasting, deep learning methods have demonstrated good accuracy in predicting complex time series. However, the excellent performance of these methods is dependent on the amount of data available. For this purpose, in this study, we propose to use time series augmentation techniques to improve the performance of deep learning methods. In this study, three new methods have been used to test the effectiveness of the proposed approach, which are: 1) Long short-term memory, 2) Convolutional network 3) Multihead self-attention mechanism. This study also uses a multi-step forecasting approach that makes it possible to predict several future points in a forecasting operation. The proposed method is applied to the actual demand data of a furniture company. The experimental results show that the proposed approach improves the forecasting accuracy of the methods used in most different prediction scenarios

Sentiment analysis is a branch of computer science and natural language processing that seeks to familiarize machines with human emotions and make them recognizable. Both sentiment analysis and sarcasm which is a sub-field of the former, seek to correctly identify the hidden positive and negative emotions of the text. The use of sarcasm on social media, where criticism can be exercised within the context of humor, is quite common. Detection of sarcasm has a special effect on correctly recognizing the polarization of an opinion, and thus not only it can help the machine to understand the text better, but also makes it possible for the respective author to get his message across more clearly. For this purpose, 8000 Persian tweets that have emotional labels and examined for the presence or absence of sarcasm have been used. The innovation of this research is in extracting keywords from sarcastic sentences. In this research, a separate classifier has been trained to identify irony of the text. The output of this classifier is provided as an added feature to the text recognition classifier. In addition to other keywords extracted from the text, emoticons and hashtags have also been used as features. Naive Bayes, support vector machines, and neural networks were used as baseline classifiers, and finally the combination of classifiers was used to identify the feeling of the text. The results of this study show that identifying the irony in the text and using it to identify emotions increases the accuracy of the results.

The Internet of Things (IoT) is a novel and widely used idea aimed at connecting objects through communication technologies. The problem of the prior technology adaptation has always been one of the most challenging issues in this area over the years. The Recognition of Prior Learning (RPL) protocol has been proposed by scientists since 2012 as a solution for IoT routing. This protocol has been utilized by many researchers and hardware companies in the field of the mentioned technology. The present study evaluates RPL behavior from the perspective of the existence of stopping conditions, crossing multiple routes from a special route (loop conditions), and how it reacts to different inputs, while presenting a modular and readable model of this protocol.

Today, information and communication technology has influenced all areas of human life, including economics and politics. In research on the effects of information and communication technology, its effects on economic risk and political risk have not been considered so far. Therefore, this study has examined the effect of ICT development on economic and political risk for three selected groups of developed, developing, and OPEC countries in the period 2007- 2019. The panel data method was used to estimate the model. Based on the model estimation results for all sample countries, the deployment of information and communication technology reduces economic and political risk; But the results are different for each group. As the development of information and communication technology in selected OPEC members and developing countries increases economic risk; But in developed countries, it reduces economic risk. Also, in OPEC member countries, no significant relationship was found between ICT and political risk, and in selected developed and developing countries, ICT increases political risk.

Quality assurance of providing remote services in cellular networks necessitates attention to significant criteria such as throughput, power consumption, and interference in these networks. Accordingly, this paper presents a framework for optimizing these criteria by assuming a limited transmission capacity for mobile nodes in a wireless cellular network as limitations in the transmission capacity often exist both in terms of hardware, battery limitations, and regulatory rules in the real world. In presenting this framework, a new idea was proposed once the existing methods were examined and their advantages and disadvantages were compared, respectively. After the formula was proved, the idea's simulation steps were performed via MATLAB. Present methods either increased the throughput by assuming unlimited transmission power or prevented some nodes from accessing the communication service. The simulation results showed that the proposed algorithm reduced the power consumption of mobile nodes in the network by a quarter in addition to increasing the throughput by 27%, and further operated in a way that no node would lose communication service

Text detection in images is one of the most important sources for image recognition. Although many researches have been conducted on text detection and recognition and end-to-end models (models that provide detection and recognition in a single model) based on deep learning for languages such as English and Chinese, the main obstacle for developing such models for Persian language is the lack of a large training data set. In this paper, we design and build required tools for synthesizing a data set of scene text images with parameters such as color, size, font, and text rotation for Persian. These tools are used to generate a large still varied data set for training deep learning models. Due to considerations in synthesizing tools and resulted variety of texts, models do not depend on synthesis parameters and can be generalized. 7603 scene text images and 39660 cropped word images are synthesized as sample data set. The advantage of our method over real images is to synthesize any arbitrary number of images, without the need for manual annotations. As far as we know, this is the first open-source and large data set of scene text images for Persian language.

Most of the previous pricing models for cloud resources which are defined based on auction suffer from high implementation complexity in real cloud environments. Therefore, the main challenge for researchers is to design dynamic pricing models that can achieve three goals: 1) low computation complexity, 2) high accuracy, and 3) high implementation simplicity in real cloud environments. CMM (Cloud Market Maker) is one of the most popular dynamic pricing models that has two advantages of computation accuracy and the possibility to implement in the real cloud environments. This model calculates the bid price based on a linear function. In designing this linear function, the parameters: buyer’s urgency, number of competitors and number of opponents are considered. Despite the advantages of this pricing function, the importance ratio of the constructor parameters of it is considered the same in various market conditions. Ignoring this issue reduces both system flexibility and computation accuracy in tangible changes in the cloud market. Therefore, the authors of this paper focus on designing a new cloud market-aware intelligent pricing system (which developed in customer side of the market) to tackle the mentioned problem. At the same time, high implementation simplicity of the proposed system should be guaranteed. For this purpose, an agent-based intelligent pricing system by combining support vector machine (SVM) and hierarchical analysis process (AHP) techniques is proposed. Simulation results show the better performance of the proposed solution which is named as DPMA in comparison to CMM.

Today, many governments around the world are increasingly leveraging advances in information and communication technologies to provide electronic services to their citizens. But estimates indicate that e-government projects will fail miserably, both in part and in whole. Incomplete formation and poor supply chain performance are the most important reasons for the failure of these projects by researchers. Due to the fact that few studies have been conducted in the field of e-government service supply chain and its performance evaluation indicators, this study has studied the e-government service supply chain in Iran by studying the e-customs system. Also, using SMART criteria and ELECTRE I technique, it has identified the performance evaluation indicators of this chain. This study shows that just as the principles of e-supply chain in the manufacturing sector have been proposed by researchers, this concept can be generalized to the public service sector. The results also show that unlike previous studies on performance appraisal indicators, e-government service supply chain indicators have significant differences with traditional service supply chains.

In this paper, problem of using the 3D beamforming method (3DBF) in a multi-input-multi-output cellular communication network (MIMO) is discussed. The network consists of a cell with multiple users, in which users are distributed based on the Poisson point (PPP) process at the cell area, which is closer to the conditions in a real mobile network. In this case, the number of users inside the cell and their location will be random. Depending on the distribution of users in the space and the difference in their distance from the base station, their elevation angles will also be different. Considering the downlink transmission and the zero-forcing (ZF) precoder in the base station, with the aim of eliminating intra cell interference, we evaluate and analyze the probability of coverage in the cell and then we obtain the best antenna tilt angle to achieve maximum probability of coverage. Using the analysis of numerical results, the accuracy of the calculations and the value of the optimal tilt angle of the antenna array are confirmed.

Stock market prediction serves as an attractive and challenging field for researchers in financial markets. Many of the models used in stock market prediction are not able to predict accurately or these models require a large amount of input data, which increases the volume of networks and learning complexity, all of which ultimately reduce the accuracy of forecasting. This article proposes a method for forecasting the stock market that can effectively predict the stock market. In this paper, the past market price is used to reduce the volume of input data and this data is placed in a regressor model.

 In this paper, we consider an uplink economy-efficient resource allocation in a multicellular virtual wireless network with a C-RAN architecture where a MNO interacts with a number of MVNOs with a predetermined business model. In each cell of this system, two types of multiple access technologies, namely OFDMA and Massive MIMO, are available for MVNO at two different prices. In this setup, we propose a multi access technology selection approach (MATSA) with the objective to reduce operating costs and maximize the profit of the MVNOs subject to a set of constraints, and formulate this resource allocation problem with the new utility function. Due to the existence of continuous and binary variables in the formulated optimization problem and also the interference between cells in data rate functions, this optimization problem will be non-convex with very high computational complexity. To tackle this problem, by applying the complementary geometric programming (CGP) and the successive convex approximation (SCA), an effective two-step iterative algorithm is developed to convert the optimization problem into two sub problems with the aim to find optimum technology selection and power consumption parameters for each user in two steps, respectively. The simulation results demonstrate that our proposed approach (MATSA) with novel utility function is more efficient than the traditional approach, in terms of increasing total EE and reducing total power consumption. The simulation results illustrate that the profit of the MVNOs is enhanced more than 13% compared to that of the traditional approach.

In this paper, a device for measuring the electric fields intensity in the environment is designed and presented in the 5G frequency band, including the frequency range of 3400 ~ 3600 MHz. This device, called the 5G electric probe, is realized by three orthogonal antennas, in connection to filter circuits and power detectors. The proposed antenna is a strip monopole antenna, and these orthogonal antennas can receive the electric fields in all directions uniformly and isotropically. The proposed filter is a coupled-line microstrip filter that has the ability to remove out-of-band signals. The proposed power detector is able to operate linearly over a wide dynamic range and convert the fields received from the antenna and filter sections to suitable DC voltages for digital processing. Finally, the designed 5G electric probe is fabricated and tested. The measurements confirm the proper operation of the probe in terms of dynamic range, accuracy, sensitivity, and the linearity and isotropicity of the received electric fields.

By widespread use of cloud computing, the need to improve performance and reduce latency in the cloud increases. One of the problems of distributed environments, especially clouds, is unbalanced load which results in reducing speed and efficiency and increasing delay in data storage and retrieval time. Various methods for load balancing in the cloud environment have been proposed, each of which has addressed the issue from its own perspective and has its advantages and disadvantages. In this research, we first provide some criteria for measuring load balance in the cloud and then examine the use of Metaheuristic methods in load balancing in the cloud environment. After introducing Metaheuristic load balancing methods, we have compared them based on the aforementioned criteria and discussed the advantages and disadvantages of each. Ant Colony Algorithms, Artificial Ant Colony, Bee Colony, Artificial Bee Colony, Bee Foraging Algorithm, Particle Swarm, Cat Swarm, Simulated Annealing, Genetic Algorithm, Tabu Search, Fish Swarm and Hybrid Algorithms and etc. examined in this research.

Today, quality is one of the most important factors in attracting customer satisfaction and loyalty to service organizations. Therefore, one of the main concerns of managers is to improve the quality of services. With the development of the Internet and the world of communications, a concept called cloud computing has expanded in the world of communications, which provides a new model for the supply, consumption and delivery of computing services. The purpose of this study is to make the optimal decision in choosing the appropriate cloud service according to the conditions of users so that they achieve the highest satisfaction. Fuzzy Delphi method, fuzzy hierarchical analysis method, fuzzy TOPSIS method and finally multi-criteria decision making method are the methods used in this research. The results of the fuzzy Delphi method show that the indicators of transparency, accessibility and reliability should be eliminated. The results of fuzzy hierarchical analysis identified the cost index as the most important index and the support index during demand as the least important index. According to the results of fuzzy TOPSIS based on the weights obtained from fuzzy hierarchical analysis, SAAS, IAAS and PAAS cloud services were ranked first to third, respectively. Using the SAAS service provides numerous benefits to employees and companies, such as reducing time and money spent on time-consuming tasks such as installing, managing, and upgrading software.

Despite the enormous volume of data and the benefits it can bring to marketing activities, it is unclear how to use it in the literature, and very few studies have been conducted in this field. In this regard, this study uses dynamic capabilities view to identify the dynamic capabilities of data-driven marketing to focus on data in the development of marketing strategies, make effective decisions, and improve efficiency in marketing processes and operations. This research has been carried out in a qualitative method utilizing the content analysis strategy and interviews with specialists. The subjects were 18 professionals in the field of data analytics and marketing. They were selected by the purposeful sampling method. This study provides data-driven marketing dynamic capabilities, including; Ability to absorb marketing data, aggregate and analyze marketing data, the ability to data-driven decision-making, the ability to improve the data-driven experience with the customer, data-driven innovation, networking, agility, and data-driven transformation. The results of this study can be a step towards developing the theory of dynamic capabilities in the field of marketing with a data-driven approach. Therefore, it can be used in training and creating new organizational capabilities to use big data in the marketing activities of organizations, to develop and improve data-driven products and services, and improve the customer experience

With the incredible development of social networks, many marketers have exploited the opportunities, and attempt to find influential people within online social networks to influence other people. This problem is known as the Influence Maximization Problem. Efficiency and effectiveness are two important criteria in the production and analysis of influence maximization algorithms. Some of researchers improved these two issues by exploiting the communities’ structure as a very useful feature of social networks. This paper aims to provide a comprehensive review of the state of the art algorithms of the influence maximization problem with special emphasis on the community detection-based approaches