Research on site laboratory data based on isolation forest and canonical correlation analysis

In order to solve the problem that a large amount of testing data about highway construction raw materials and engineering entities accumulated in the site laboratory can only be used for the current construction management, it cannot extract the effective information contained in the data, a method based on isolated forest, principal component analysis, and canonical correlation analysis was proposed. Firstly, the isolation forest algorithm was used to eliminate the abnormal data, and then the principal component analysis (PCA) algorithm was used to obtain the main indicators representing the quality of highway raw materials. Finally, the canonical correlation analysis (CCA) method was used to study the correlation between the two groups of variables: the construction raw materials testing data and the engineering entity testing data. Through the above steps, the information contained in the test data of the site laboratory can be effectively extracted, so as to provide effective suggestions for the construction material selection.