Population loss early warning method, device and equipment based on multi-dimensional feature fusion learning

Abstract

The invention discloses a population loss early warning method, device and equipment based on multi-dimensional feature fusion learning, and the method comprises the steps: determining a multi-dimensional feature set of a person, and the multi-dimensional features are features of a target person in a plurality of preset dimensions related to population loss; taking the multi-dimensional feature set as input of a pre-established Stacking integration model to obtain a loss analysis result of the target personnel; wherein the Stacking integrated model comprises a plurality of learning devices on a first layer and a Stacking model on a second layer, the plurality of learning devices respectively carry out K-fold cross validation on features in the multi-dimensional feature set, and a prediction result is used as input of the Stacking model to obtain a loss analysis result; the Stacking integrated model is obtained through training based on a training sample corresponding to the personnel and a loss label corresponding to the training sample, the training sample has features with the same dimension as the multi-dimensional feature set, and the loss label is used for representing whether the personnel are lost or not. Therefore, the multi-dimensional feature fusion integrated model can be adopted to perform pre-judgment analysis on the population loss, and the population loss behavior is pre-warned in advance, so that decision service is provided for the region.

Description

technical field [0001] This document relates to the field of computer technology, in particular to a population loss early warning method, device and equipment based on multi-dimensional feature fusion learning. Background technology [0002] In the trend of social development, urban population loss will seriously affect urban development. Population is the basis of production and consumption. Regional population loss will not only lead to the decline of regional labor force, but also the decline of regional consumption capacity. At present, most of the early warning research on population loss at home and abroad is in theoretical research, paying attention to the mechanism and theoretical analysis of the whole process of population loss, and testing the influencing factors in the early warning process of population loss. There are two main methods for early warning of mainstream population loss, namely qualitative analysis and quantitative analysis. Qualitative analysis mainly e...

AI Technical Summary

Problems solved by technology

[0007] This manual provides a population loss early warning method based on multi-dimensional feature fusion learning to solve the problem of being unable to objectively provide early warning of population loss

Images