The COVID-19 pandemic has had a devastating impact on the world, and identifying high-risk areas is essential for preventing the spread of the virus. In this paper, the authors propose a novel regression method called Regular Linear Kernel Regression (RLKR) for COVID-19 high-risk areas exploration. RLKR is a regularized version of linear kernel regression, which is a standard machine learning method for regression tasks. The authors show that RLKR can be used to identify high-risk areas for COVID-19 by exploiting the spatial correlation of the virus. They also prove that RLKR is consistent under two mild assumptions. The authors evaluate RLKR on a real-world dataset of COVID-19 cases in China. They show that RLKR is able to identify high-risk areas with high accuracy. They also show that RLKR is more accurate than other methods for identifying high-risk areas. The authors conclude that RLKR is a promising new method for COVID-19 high-risk areas exploration. They suggest that RLKR could be used to improve the effectiveness of public health interventions during the COVID-19 pandemic.