Travel generation prediction method, system and device based on gradient boosting decision tree

Abstract

The invention belongs to the field of population travel generation and prediction, particularly relates to a travel generation and prediction method, system and device based on a gradient boosting decision tree, thereby solving problems that an existing travel generation method cannot truly reflect the nonlinear relationship between an input value and prediction, the model inspection calculation amount is large, and a result is not visual. The method comprises the steps: extracting independent variables of current travel generation data of each traffic cell in a to-be-predicted area, and performing normalization processing; obtaining a prediction value of each current traffic cell of the to-be-predicted area through a travel generation prediction model; and carrying out reverse normalization on the prediction values to obtain prediction travel generation data of each current traffic cell of the to-be-predicted area. According to the method, the nonlinear relation between original inputand output can be accurately reflected, the square error principle is used for searching the minimum division characteristic and division points, redundant variables are automatically ignored, the manual screening process of the variables is omitted, and high precision and robustness are achieved.

Description

technical field [0001] The invention belongs to the field of population trip generation prediction, and in particular relates to a trip generation prediction method, system and device based on a gradient lifting decision tree. Background technique [0002] The interactive relationship between urban traffic and urban land use determines that different types and intensities of social activities will result from different land use layout forms and intensities, thus determining the traffic distribution and distribution in different regions. Correspondingly, the functional efficiency of the transportation system also directly affects the surrounding land price, land rent and popularity, and affects whether the surrounding land functions are fully realized. Therefore, in the transportation planning, it is necessary to study the relationship between urban land use and traffic in depth, and the traffic trip rate is one of the important indicators that intuitively reflect this relati...

AI Technical Summary

Problems solved by technology

[0006] In order to solve the above-mentioned problems in the prior art, that is, the existing travel generation method cannot truly reflect the nonlinear relationship between the input value and the prediction, and the model checking has a large amount of calculation and the result is not intuitive, the present invention provides a gradient-based A trip generation prediction method for boosting a decision tree that includes:

Images