Research on Characteristic Identification of Relatively Poor Migrant Workers Based on Random Forest Model

DOI

10.2991/assehr.k.210916.013How to use a DOI?

Keywords

Migrant workers, Relative poverty, Double bounds method, Random forest

Abstract

After China has eliminated absolute poverty and entered the post-poverty era, the anti-poverty cause has evolved to focus on relative poverty, which has the characteristics of relativity, multidimensionality, and dynamics. In particular, migrant workers have become the main group of relative poverty. Its characteristics such as strong mobility, difficulty in continuously increasing income, and lack of endogenous motivation make it more difficult to identify and manage relative poverty. Therefore, this paper uses the double-boundary method to identify and analyze the relatively poor migrant workers. It is found that: (1) Compared with the one-dimensional income level as the standard, the multi-dimensional poverty identification system can better reflect the poverty status of migrant workers; (2) The migrant workers showed a higher incidence of poverty in the highest education level, children’s school attendance, five social insurance and one housing fund and other indicators; (3) The random forest algorithm was used to construct the relative poverty identification model of migrant workers, with an accuracy rate of 98.8%. Meanwhile, the model reflected that education and employment dimensions should be emphasized in the identification process. Based on the research conclusions, this paper puts forward corresponding policy suggestions for China to deal with the relative poverty of migrant workers and puts forward further research directions.

Copyright

© 2021, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Lilu Sun
AU  - Xin Ma
AU  - Yuxin Ding
PY  - 2021
DA  - 2021/09/16
TI  - Research on Characteristic Identification of Relatively Poor Migrant Workers Based on Random Forest Model
BT  - Proceedings of the 2021 International Conference on Social Science:Public Administration, Law and International Relations (SSPALIR 2021)
PB  - Atlantis Press
SP  - 85
EP  - 92
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.210916.013
DO  - 10.2991/assehr.k.210916.013
ID  - Sun2021
ER  -