Time map and predictors of on-spot emotional responses of Chinese people during COVID-19 outbreak: From January 27 to February 20, 2020

Background: The impact of the Coronavirus Disease 2019 (COVID-19) pandemic on people's mental health has been gradually revealed. Emotional responses under this pandemic have been a focus. This study aimed to observe the time map and predictors of transient daily emotions (on-spot emotional responses) during the COVID-19 outbreak in Chinese people. Methods: A total of 133,027 Chinese people from all cities and areas of China took part in this 25-day online investigation from January 27 to February 20, 2020. Hierarchical regression was carried out. Results: Chinese people reported mild positive and negative emotions during COVID-19, and the trend was that negative emotions declined and happy emotion increased in a curvilinear style. Females reported stronger on-spot negative emotions. Young and unmarried people, and individuals with lower risk perception reported lower negative emotions and higher happiness, while people with postgraduate or higher education, medical staff and patients, and individuals coming from areas with more confirmed cases, experienced stronger positive and negative emotions. In the regression model, risk predictors of negative emotions were being female, being older in age, having postgraduate or higher education, living in a city with more confirmed cases, being a patient or medical staff, and having higher risk perception, while protective predictors of negative emotions were being unmarried, having high-school or college education, and later investigating date. Notably, been married was negative predictor of happiness. Conclusions: This large sample, 25-day successive online investigation is among the first to reveal the time map of on-spot emotions in Chinese people under the COVID-19 pandemic. The risk and protective predictors of on-spot negative emotions help to recognize vulnerable people under pandemic conditions and further develop more targeted early interventions for future crises. © 2021 The Authors

Age Prediction of Death in India' CoVID-19 Pandemic using Machine Learning Methods

CoVID-19 is a widespread pandemic all over the world. It is important to give in-depth research on it. In this work, we aim to predict the age of death in India's CoVid-19. We conduct this work using two machine learning methods, including AdaBoost and Random Forest. Then, we also apply feature selection to speed up the training processing, using random selection, PCA, SVD, and correlation. Between those two methods, we find that the AdaBoost is more suitable for age prediction, and the correlation method, with little performance decreasing, achieves 3 times speed up during training. © 2021 IEEE.

Revealing Drivers of Haze Pollution by Explainable Machine Learning

Many places on earth still suffer from a high level of atmospheric fine particulate matter (PM2.5) pollution. Formation of a particulate pollution event or haze episode (HE) involves many factors, including meteorology, emissions, and chemistry. Understanding the direct causes of and key drivers behind the HE is thus essential. Traditionally, this is done via chemical transport models. However, substantial uncertainties are introduced into the model estimation when there are significant changes in the emissions inventory due to interventions (e.g., the COVID-19 lockdown). Here we applied a Random Forest model coupled with a Shapley additive explanation algorithm, a post hoc explanation technique, to investigate the roles of major meteorological factors, primary emissions, and chemistry in five severe HEs that occurred before or during the COVID-19 lockdown in China. We discovered that, in addition to the high level of primary emissions, PM2.5 in these haze episodes was largely driven by meteorological effects (with average contributions of 30-65 mu g m(-3) for the five HEs), followed by chemistry (similar to 15-30 mu g m(-3)). Photochemistry was likely the major pathway of formation of nitrate, while air humidity was the predominant factor in forming sulfate. Our results highlight that the machine learning driven by data has the potential to be a complementary tool in predicting and interpreting air pollution.

Practice and thinking of online teaching during epidemic period

The COVID-19 suddenly came and spread worldwide, the education department advocates the suspension of classes, but it keeps using the network platform to teach and learn. The establishment of effective online education methods and mechanisms plays an important role in improving teachers' teaching and students' learning. Based on the thinking of online teaching methods during the epidemic period, this paper puts forward a teaching concept centered on "learning effect". Use Nail, Smart Rain, Python 123, MOCC and other online teaching platforms. In order to improve the teaching effect in a virtuous circle, we adopt the teaching design of stages before, during and after class, and diversified teaching methods and multiple dimensional teaching evaluation through encouragement and punishment. This method has achieved good teaching effect in the process of teaching practice, and can be extended to other online teaching courses for a long time. © 2020 IEEE.

Construction and application of the information system for preview and triage during novel coronavirus pneumonia epidemic time in stomatological hospitals

Objective: To construct an information system for preview and triage during novel coronavirus pneumonia (NCP) epidemic time in stomatological hospitals and put it to clinical use. Methods: To sort out and formulate the preview and triage process of stomatological hospitals during the epidemic, we constructed B/S development platform with Jeesite open source framework, sort outed and developed relevant input and query functions. ActiveX widget was used to connect the hardware part with the ID card reader through the browser. At the initial stage of the system put online, the parallel mode of manual registration and system registration was used to verify the actual effect of the system. Results: After the system went online, the triage data of 35 patients were analyzed. The average triage time was shortened from (90.82±31.85) seconds to (38.97±12.14) seconds, with an efficiency increase of about 125%. Results: The system's performance can meet the actual business requirements, reduce the risk of cross infection, improve the preview and triage efficiency, and improve patients' medical experience. Meanwhile it ensures the accuracy and safety of the information record of people seeking medical service.