درخت حوزه‌های تخصصی

The introduction of cloud computing techniques revolutionized the current of information processing and storing. Cloud computing as a competitive edge provides easy and automated access to the vast ocean of resources through standard network mechanisms to businesses and organizations. Due to the vast diversity of service providers and their respective variety of available services with different qualities, top managements often face difficulty for choosing the best available option. So, considering the growing significance of the mentioned issue, this study aims to identify and rank contributing factors in selection of cloud service providers. In that attempt, this research approaches its goal by going through three major phases. Firstly, in phase one, prior studies are reviewed for extracting related elements of selection. Secondly, by employing Fuzzy Delphi method and obtaining results by interviewing experts in this field such as IT managers and technicians, this study tries to finalize the list of contributing factors. Lastly, by utilizing Fuzzy best-worst multi-criteria decision-making method, which is one of the most recent techniques employed to statistically rank variables, this research introduces a list of vital factors for cloud service selection. Based on the findings of this study, there are five major categories involved in the selection process which are: performance, security, data management, personal data protection and environmental-organizational. The finalized result of ranking shows that, performance related factors such as accessibility, response time and capacity are the first priority. The runner-up is security with reliability and governance. Environmental-organizational variables lands in the third place by considering rental and network costs.

Due to the huge amount of data published on the Web, the Web search process has become more difficult, and it is sometimes hard to get the expected results, especially when the users are less certain about their information needs. Several efforts have been proposed to support exploratory search on the web by using query expansion, faceted search, or supplementary information extracted from external knowledge resources. However, these solutions are not well explored for the general web search in an open-domain setting. In addition, they mostly focus on supporting search in content expressed in English and Latin based languages. In this research, we propose a fully automated approach that aims to support exploratory search over the Arabic web content. It exploits the Arabic version of Wikipedia to extract complementary information that supports visual representation and deeper exploration of the search engine's results. Key Wikipedia entities are extracted from the text snippets produced by the search engine in response to the user's query. Entities are then filtered and ranked by using a novel ranking algorithm that extends the conventional PageRank algorithm. Finally, a graph is built and presented to the user to visually represent highly ranked topics and their relationships. The proposed approach was realized by developing ArabXplore, a system that integrates with the web browser to support the web search process by executing our approach in query time. It was assessed over a dataset of 100 Arabic search queries covering different domains, and results were assessed and rated by human subjects. The underlying ranking algorithm was also compared with the conventional PageRank.

In recent privacy has emerged as one of the major concerns of deep learning, since it requires huge amount of personal data. Medical Image Analysis is one of the prominent areas where sensitive data are shared to a third party service provider. In this paper, a secure deep learning scheme called Metamorphosed Learning (MpLe) is proposed to protect the privacy of images in medical image analysis. An augmented convolutional layer and image morphing are two main components of MpLe scheme. Data providers morph the images without privacy information using image morphing component. The human unrecognizable image is then delivered to the service providers who then apply deep learning algorithms on morphed data using augmented convolution layer without any performance penalty. MpLe provides sturdy security and privacy with optimal computational overhead. The proposed scheme is experimented using VGG-16 network on CIFAR dataset. The performance of MpLe is compared with similar works such as GAZELLE and MiniONN and found that the MpLe attracts very less computational and data transmission overhead. MpLe is also analyzed for various adversarial attack and realized that the success rate is as low as . The efficiency of the proposed scheme is proved through experimental and performance analysis.