ABSTRACT
While the emergence of COVID-19 [1] has put major cloud service providers around the world to the test, the pandemic has also provided a strong impetus for the adoption and deployment of cloud computing: the transition to a remote workforce, entertainment, e-commerce, and especially remote education have affected the cloud industry and how providers are responding to the sudden and significant increase in demand for cloud solutions and services. Obviously, while highlighting the robustness of the public cloud, the pandemic-induced situation also highlights several important research challenges that need to be addressed. This paper presents a multi-source based analysis for the identification of cloud computing research challenges as part of the road mapping methodology followed in the HUB4CLOUD project. The analysis consists of an in-depth study of several sources including analysis of the international context, analysis of academic venues, interviews with relevant stakeholders and existing funded projects. The paper also provides an overview of the main research topics identified and proposes next steps for the utilization of these finding in the development of a Cloud Computing research roadmap. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
ABSTRACT
The ongoing COVID-19 pandemic has overloaded current healthcare systems, including radiology systems and departments. Machine learning-based medical imaging diagnostic approaches play an important role in tracking the spread of this virus, identifying high-risk patients, and controlling infections in real-time. Researchers aggregate radiographic samples from different data sources to establish a multi-source learning scheme to mitigate the insufficiency of COVID-19 samples from individual hospitals, especially in the early stage of the disease. However, data heterogeneity across different clinical centers with various imaging conditions is considered a significant limitation in model performance. This paper proposes a contrastive learning scheme for the automatic diagnosis of COVID-19 to effectively mitigate data heterogeneity in multi-source data and learn a robust and generalizable model. Inspired by advances in domain adaptation, we employ contrastive training objectives to promote intra-class cohesion across different data sources and inter-class separation of infected and non-infected cases. Extensive experiments on two public COVID-19 CT datasets demonstrate the effectiveness of the proposed method for tackling data heterogeneity problems with boosted diagnosis performance. Moreover, benefiting from the contrastive learning framework, our method can be generalized to solve data heterogeneity problems under a broader multi-source learning setting. © 2021 IEEE
ABSTRACT
Urban vitality analysis is an important means to measure the attractiveness and cohesion of urban development, which is of great significance to urban operation situation awareness, fine governance and planning design. Taking China Singapore Tianjin Eco-city as the research area, based on multisource temporal-spatial data from January 2020 to February 2021 provided by the smart city data aggregation platform, such as the data of urban transportation, catering waste recycling volume, books borrowed and returned, this paper analyzed the urban vitality index from five aspects of society, transportation, commerce, culture-education and tourism. After spatializing data into geographic grids, method EW-TOPSIS (Entropy Weighted Technique for Order Preference by Similarity to an Ideal Solution) has been utilized to calculate the comprehensive vitality index. The results showed that: from temporal perspective, the vitality of each index was significantly different due to the impact of Corona Virus Disease 2019(COVID-I9) and seasons;the comprehensive vitality decreased sharply to the lowest at the beginning of the COVID-19 while it increased rapidly and then stabilized after the weakening of the prevention and control policies;the comprehensive vitality value of Eco-city reached the peak of the whole year during the National Day Golden Week affected by tourism vitality. From spatial perspective, the distribution of Eco-city vitality level is unbalanced, showing a trend of outward diffusion centered on the urban starting area and the main tourist attractions. © 2021 IEEE.