A Multi-dimensional Relative Poverty Measurement and Early Warning Model in the Yangtze River Economic Belt Based on Adaboost
Ensuring that there is no large-scale return to poverty is an important aspect of relative poverty.And accurately identifying relative poverty groups is a prerequisite for achieving long-term poverty governance.This article constructs a multidimensional poverty indicator system based on the 2020 China Family Panel Studies database.The A-F double boundary method is used to measure the relative poverty situation of households in 11 provinces and cities in the Yangtze River Economic Belt region.And a poverty label is added to each household.The Adaboost algorithm is used to establish a relative poverty household identification model in the Yangtze River Economic Belt region.The results show that the accuracy,recall,precision,and F1 score of the model for predicting relative poverty reached 99.66%,100.00%,99.43%and 99.71%of urban data and 99.09%,100.00%,97.73%and 98.8%of rural data.Respectively,it demonstrates the good generalization ability of the model.By analyzing the importance of each characteristic variable,it was found that education,loan rejection,healthcare,income,and household assets have a greater impact on relative poverty in households.
Yangtze River Economic Beltrelative poverty warningAdaboostA-F method
万方数据
维普
