Facility site selection method based on graph convolutional neural network

Abstract

The invention discloses a facility site selection method based on a graph convolutional neural network. The method comprises the steps: carrying out the prediction and evaluation of all candidate position points through an inputted to-be-selected facility type, the spatial layout of built-up facilities in a current target range, and observable human activity intensity information, and selecting anoptimal position as a site selection result for output. Technically, modeling is carried out on the position relationship and the environmental characteristics of urban facilities through a model with strong spatial relationship depicting capability, namely a graph convolutional neural network, and objective and accurate urban facility candidate position prediction and evaluation are realized ina data driving manner by taking observed urban human activity data as a training basis of the model. The defects that target constraints are too subjective and the computing power is insufficient in the urban facility site selection problem are overcome, and the decision quality is improved. And an intelligent and objective facility site selection solution can be provided for a high-density and multi-attribute-dimension urban space.

Description

Technical field [0001] The present invention relates to the field of data processing, and in particular to a facility location method based on graph convolutional neural network. Background technique [0002] The rapid development of information and communication technology and the explosive emergence of geographic information containing location marks and attribute data provide a basis for people to understand the socio-economic characteristics of public facilities in cities. [0003] Current urban facility location selection is mostly based on macro-planning decision-making or space optimization algorithms. The basic idea is to use the location relationship of the established facilities and the corresponding functional conditions to define constraints and objective functions, and solve the planning equations to determine the space of the proposed facility position. For example, to build a supermarket in a certain area to serve all users in that area, it is necessary to consider ...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a facility site selection method based on graph convolutional neural network, thereby solving the aforementioned problems existing in the prior art

Images