County-level unit time sequence GDP prediction method based on deep learning

Abstract

The invention provides a method for predicting county-level unit timing sequence GDP based on deep learning. The method comprises the steps that firstly, collecting and managing multi-source remote sensing data, county-level vector boundary data and annual county-level GDP data of a time sequence in a research area based on a GEE platform; then, performing data processing based on a GEE platform,providing a feature extraction method based on histogram statistics, packaging features into tensors needed by deep learning, and dividing the data into a training set and a verification set accordingto time; establishing a deep learning system structure based on a convolutional neural network, inputting the data feature tensor into the network, performing training by using historical data, and storing the trained model; and finally, inputting the verification actual data into the model to obtain a prediction result. According to the method provided by the invention, the prediction precisioncan be improved, the labor cost is saved, and related departments and governments can be helped to count accurate social and economic data, so that industrial assistance is provided in a targeted manner and policies are formulated.

Description

technical field [0001] The invention relates to the technical field of surface parameter inversion of remote sensing data, in particular to a method for predicting time-series GDP of county-level units based on deep learning. Background technique [0002] Gross domestic product (GDP), as the most important variable in the analysis of economic growth, represents the final results of the production activities of all resident units in a country (or region) within a certain period of time, and plays an important role in evaluating the economic performance of a country or region. important role. However, the existing serious problem is that GDP data is often mismeasured, especially in developing countries, where the proportion of economic activities is not high, the degree of economic integration in the region is low, and the infrastructure is relatively weak. These factors make it difficult for relevant departments to Accurate statistics on socio-economic data, poverty levels, ...

AI Technical Summary

Problems solved by technology

Some scholars took 5 developing countries as objects, based on convolutional neural network, used nighttime light data to predict the poverty index in the region, and found that the model explained up to 75% of the changes in econo

Images