Since December 2019, the COVID-19 outbreak has resulted in countless deaths and has harmed all facets of human existence. COVID-19 has been designated an epidemic by the World Health Organization ...(WHO), which has placed a tremendous burden on nearly all countries, especially those with weak health systems. However, Deep Learning (DL) has been applied in several applications and many types of detection applications in the medical field, including thyroid diagnosis, lung nodule recognition, fetal localization, and detection of diabetic retinopathy. Furthermore, various clinical imaging sources, like Magnetic Resonance Imaging (MRI), X-ray, and Computed Tomography (CT), make DL a perfect technique to tackle the epidemic of COVID-19. Inspired by this fact, a considerable amount of research has been done. A Systematic Literature Review (SLR) has been used in this study to discover, assess, and integrate findings from relevant studies. DL techniques used in COVID-19 have also been categorized into seven main distinct categories as Long Short Term Memory Networks (LSTM), Self-Organizing Maps (SOMs), Conventional Neural Networks (CNNs), Generative Adversarial Networks (GANs), Recurrent Neural Networks (RNNs), Autoencoders, and hybrid approaches. Then, the state-of-the-art studies connected to DL techniques and applications for health problems with COVID-19 have been highlighted. Moreover, many issues and problems associated with DL implementation for COVID-19 have been addressed, which are anticipated to stimulate more investigations to control the prevalence and disaster control in the future. According to the findings, most papers are assessed using characteristics such as accuracy, delay, robustness, and scalability. Meanwhile, other features are underutilized, such as security and convergence time. Python is also the most commonly used language in papers, accounting for 75% of the time. According to the investigation, 37.83% of applications have identified chest CT/chest X-ray images for patients.
•Presenting a broad review of the existing issues related to DL methods in COVID-19.•Presenting a systematic overview of the existing methods for DL-COVID-19.•Presenting an explanation of the important methods in DL combining COVID-19.•Investigating each approach that referred to Dl-COVID-19 with various properties.•Outlining the key zones that can improve the mentioned techniques in the future.
PurposeThe main goal of this paper is to study the cloud service discovery mechanisms. In this paper, the discovery mechanisms are ranked in three major classes: centralized, decentralized, and ...hybrid. Moreover, in this classification, the peer-to-peer (P2P) and agent-based mechanisms are considered the parts of the decentralized mechanism. This paper investigates the main improvements in these three main categories and outlines new challenges. Moreover, the other goals are analyzing the current challenges in a range of problem areas related to cloud discovery mechanisms and summarizing the discussed service discovery techniques.Design/methodology/approachSystematic literature review (SLR) is utilized to detect, evaluate and combine findings from related investigations. The SLR consists of two key stages in this paper: question formalization and article selection processes. The latter includes three steps: automated search, article selection and analysis of publication. These investigations solved one or more service discovery research issues and performed a general study of an experimental examination on cloud service discovery challenges.FindingsIn this paper, a parametric comparison of the discovery methods is suggested. It also demonstrates future directions and research opportunities for cloud service discovery. This survey will help researchers understand the advances made in cloud service discovery directly. Furthermore, the performed evaluations have shown that some criteria such as security, robustness and reliability attained low attention in the previous studies. The results also showed that the number of cloud service discovery–related articles rose significantly in 2020.Research limitations/implicationsThis research aimed to be comprehensive, but there were some constraints. The limitations that the authors have faced in this article are divided into three parts. Articles in which service discovery was not the primary purpose and their title did not include the related terms to cloud service discovery were also removed. Also, non-English articles and conference papers have not been reviewed. Besides, the local articles have not been considered.Practical implicationsOne of the most critical cloud computing topics is finding appropriate services depending on consumer demand in real-world scenarios. Effective discovery, finding and selection of relevant services are necessary to gain the best efficiency. Practitioners can thus readily understand various perspectives relevant to cloud service discovery mechanisms. This paper's findings will also benefit academicians and provide insights into future study areas in this field. Besides, the drawbacks and benefits of the analyzed mechanisms have been analyzed, which causes the development of more efficient and practical mechanisms for service discovery in cloud environments in the future.Originality/valueThis survey will assist academics and practical professionals directly in their understanding of developments in service discovery mechanisms. It is a unique paper investigating the current and important cloud discovery methods based on a logical categorization to the best of the authors’ knowledge.
Cloud computing as a new computing paradigm has a great capacity for storing and accessing the remote data and services. Presently, many organizations decide to reduce the burden of local resources ...and support them by outsourcing the resources to the cloud. Typically, scalable resources are provided as services over the Internet. The way of choosing appropriate services in the cloud computing is done by determining the different Quality of Service (QoS) parameters to perform optimized resource allocation. Therefore, service composition as a developing approach combines the existing services to increase the number of cloud applications. Independent services can be integrated into complex composited services through service composition. In this paper, a new hybrid method is proposed for efficient service composition in the cloud computing. The agent-based method is also used to compose services by identifying the QoS parameters and the particle swarm optimization (PSO) algorithm is employed for selecting the best services based on fitness function. The simulation results have shown the performance of the method in terms of reducing the combined resources and waiting time.
Recently, a new technology topic has been known as the Internet of Things (IoT), where all devices like smartphones, smart TVs, medical and healthcare ones, and home appliances have been applied for ...data generating. Due to the variety of services, the numerous service composition problems, mostly related to the Quality-of-Service (QoS) parameters, are recognized in the IoT domain. Since this issue is an NP-hard obstacle, different metaheuristic approaches have been utilized up until now to solve it. Many varieties of services can be brought into the IoT, depending on users' demands. In this research, we have proposed an effective way based on a hidden Markov model (HMM) and an ant colony optimization (ACO) to answer the service composition issue by enhancing the QoS. The HMM has been trained to predict QoS. The emission and transition matrices have been improved using the Viterbi algorithm. We have executed the QoS estimation using the ACO algorithm and found a suitable path. The outcomes have illustrated the efficacy of the introduced method regarding availability, response time, cost, reliability, and energy consumption compared to the previous methods.
Purpose
The aim of this paper is to provide a comprehensive and detailed review of the state-of-the-art mechanisms of knowledge sharing in the education field as well as directions for future ...research.
Design/methodology/approach
In the current study, a systematic literature review until June 2017 is presented, which has been on the education’s mechanisms of knowledge sharing. The authors identified 237 papers, which are reduced to 71 primary studies through the paper selection process.
Findings
By providing the state-of-the-art information, the challenges and issues, this survey will directly support academics, researchers and practicing professionals in their understanding of knowledge sharing developments in education.
Research limitations/implications
There are several limitations in this study. First, this study limited the search for articles to Google scholar and four online databases. There might be other academic journals, which may be able to provide a more comprehensive picture of the articles related to the knowledge sharing in education. Second, non-English publications were excluded from this study. The authors believe research regarding the application of knowledge sharing techniques have also been discussed and published in other languages. In addition, more studies need to be carried out using other methodologies such as interviews.
Originality/value
The paper presents a comprehensive structured literature review of the articles’ mechanisms of knowledge sharing in the education field. The paper’s findings can offer insights into future research needs.
This paper proposes a new intrusion detection system (IDS) based on a combination of a multilayer perceptron (MLP) network, and artificial bee colony (ABC) and fuzzy clustering algorithms. Normal and ...abnormal network traffic packets are identified by the MLP, while the MLP training is done by the ABC algorithm through optimizing the values of linkage weights and biases. The CloudSim simulator and NSL-KDD dataset are used to verify the proposed method. Mean absolute error (MAE), root mean square error (RMSE), and the kappa statistic are considered as evaluation criteria. The obtained results have indicated the superiority of the proposed method in comparison with state-of-the-art methods. Keywords: Intrusion detection system, Cloud computing, Neural network, Artificial bee colony, Fuzzy clustering
In the information systems, customer relationship management (CRM) is the overall process of building and maintaining profitable customer relationships by delivering superior customer value and ...satisfaction with the goal of improving the business relationships with customers. Also, it is the strongest and the most efficient approach to maintaining and creating the relationships with customers. However, to the best of our knowledge and despite its importance, there is not any comprehensive and systematic study about reviewing and analyzing its important techniques. Therefore, in this paper, a comprehensive study and survey on the state of the art mechanisms in the scope of the CRM are done. It follows this goal by looking at five categories in which CRM plays a significant role: E-CRM, knowledge management, data mining, data quality and, social CRM. In each category, a couple of studies are presented and determinants of CRM are described and discussed. The major development in these five categories is reviewed and the new challenges are outlined. Also, a systematic literature review (SLR) in each of these five categories is provided. Furthermore, insights into the identification of open issues and guidelines for future research are provided.
•Providing a systematic overview of the existing techniques in the field of CRM.•Highlighting the advantages and disadvantages of the existing techniques in the field of CRM.•Exploring some of the main challenges in the field of CRM.•Dividing the CRM techniques into five categories.•Outlining the key areas where future research can improve the use of CRM.
•Presenting a resource discovery approach to address multi-attribute and range queries.•Modeling the behaviors of proposed approach.•Analyzing the soundness, completeness, and consistency as a model ...checking problem.•Translating the design model into a verifiable formal model.•Implementing the behavior models by ArgoUML tool and the NuSMV model checker.
Grid computing is the federation of resources from multiple locations to facilitate resource sharing and problem solving over the Internet. The challenge of finding services or resources in Grid environments has recently been the subject of many papers and researches. These researches and papers evaluate their approaches only by simulation and experiments. Therefore, it is possible that some part of the state space of the problem is not analyzed and checked well. To overcome this defect, model checking as an automatic technique for the verification of the systems is a suitable solution. In this paper, an adopted type of resource discovery approach to address multi-attribute and range queries has been presented. Unlike the papers in this scope, this paper decouple resource discovery behavior model to data gathering, discovery and control behavior. Also it facilitates the mapping process between three behaviors by means of the formal verification approach based on Binary Decision Diagram (BDD). The formal approach extracts the expected properties of resource discovery approach from control behavior in the form of CTL and LTL temporal logic formulas, and verifies the properties in data gathering and discovery behaviors comprehensively. Moreover, analyzing and evaluating the logical problems such as soundness, completeness, and consistency of the considered resource discovery approach is provided. To implement the behavior models of resource discovery approach the ArgoUML tool and the NuSMV model checker are employed. The results show that the adopted resource discovery approach can discovers multi-attribute and range queries very fast and detects logical problems such as soundness, completeness, and consistency.
Wireless sensor network (WSN) is one of the interesting issues in the information technology domain. It is used in various fields such as medicine, agriculture, meteorology, etc. that eases lots of ...difficult tasks to do but have some challenges too. Deployment is the greatest challenge in WSN that affects other features like coverage, connectivity, energy efficient and lifetime. Despite the importance of the deployment problem in WSN, to the best of our knowledge, there isn’t any systematic literature review to give us systematical analyses the state-of-the mechanisms in this field. Therefore, this study reviewed the deployment mechanisms which have been used in WSN systematically. The deployment mechanisms can be classified into two main categories: deterministic and nondeterministic. Also, this study represents a comparison of the important techniques of the selected articles in each category to offer a guideline for further studies and new challenges. We also have noted some open issues for future research.