Visual analysis of impact factors of forest pests and diseases

In the field of forest pest and disease research, researchers have combined the experience and data accumulated over many years and conducted long-term and systematic observations of the research object; they have used regression analysis to determine the factors that affect the occurrence of pests and diseases. This traditional approach is time-consuming and highly dependent on expert experience. In this paper, we propose a multicombination multivariable linear regression model to quantitatively describe the multiple linear combinations of relationship between multiple independent variables and a single dependent variable. Based on this model and a data flow model combined with statistical principles and visualization techniques, we propose a multicombination multivariable linear regression visual analysis method to assist researchers in quickly assessing the correlations between the disease indexes of forest diseases and pests and the factors that may affect the pest and disease occurrences. Based on this approach, a multicombination multivariable linear regression visual analysis system was designed and implemented, and the cases of a given forest pest and disease data set were analyzed. It is shown that the multicombination multivariable linear regression visual analysis method can effectively assist researchers in quickly understanding pest and disease data, determining impact factors, and finding relevant laws. Graphic abstract