Detecting the period of a disease is of great importance to building information management capacity in disease control and prevention. This paper aims to optimize the disease surveillance process by further identifying the infectious or recovered period of flu cases through social media. Specifically, this paper explores the potential of using public sentiment to detect flu periods at word level. At text level, we constructed a deep learning method to classify the flu period and improve the classification result with sentiment polarity. Three important findings are revealed. Firstly, bloggers in different periods express significantly different sentiments. Blogger sentiments in the recovered period are more positive than in the infectious period when measured by the interclass distance. Secondly, the optimized disease detection process can substantially improve the classification accuracy of flu periods from 0.876 to 0.926. Thirdly, our experimental results confirm that sentiment classification plays a crucial role in accuracy improvement. Precise identification of disease periods enhances the channels for the disease surveillance processes. Therefore, a disease outbreak can be predicted credibly when a larger population is monitored. The research method proposed in our work also provides decision making reference for proactive and effective epidemic control and prevention in real time.
Accurately predicting the number of severe and critical COVID-19 patients is critical for the treatment and control of the epidemic. Social media data have gained great popularity and widespread application in various research domains. The viral-related infodemic outbreaks have occurred alongside the COVID-19 outbreak. This paper aims to discover trustworthy sources of social media data to improve the prediction performance of severe and critical COVID-19 patients. The innovation of this paper lies in three aspects. First, it builds an improved prediction model based on machine learning. This model helps predict the number of severe and critical COVID-19 patients on a specific urban or regional scale. The effectiveness of the prediction model, shown as accuracy and satisfactory robustness, is verified by a case study of the lockdown in Hubei Province. Second, it finds the transition path of the impact of social media data for predicting the number of severe and critical COVID-19 patients. Third, this paper provides a promising and powerful model for COVID-19 prevention and control. The prediction model can help medical organizations to realize a prediction of COVID-19 severe and critical patients in multi-stage with lead time in specific areas. This model can guide the Centers for Disease Control and Prevention and other clinic institutions to expand the monitoring channels and research methods concerning COVID-19 by using web-based social media data. The model can also facilitate optimal scheduling of medical resources as well as prevention and control policy formulation.
Ambient occlusion (AO) algorithm is a computer graphics rendering algorithm for the global background. As a crucial part of the photorealistic rendering technology, it enables users to feel the visual separation effect between 3D objects, adding more realism and hierarchy to the rendering. With the vigorous development of artificial intelligence and deep learning technology, the widespread application of convolutional neural networks provides a new research direction and solution for the further innovation of this algorithm. However, the application of convolutional neural network in the ambient light occlusion algorithm is still immature, and there are still problems such as inaccurate rendering details. The network structure and loss function cannot express the AO value well, which also makes the rendering effect unsatisfactory. In view of the above existing problems, this paper optimizes and improves on the basis of predecessors, and proposes a residual neural network AO algorithm with higher efficiency and better rendering effect, referred to as ERAO.