摘要
为巩固拓展脱贫攻坚成果同乡村振兴有效衔接,打好实施乡村振兴战略的基础,文章选取2020年中国乡村振兴综合调查(CRRS)数据为样本,以可持续生计理论为指导构建返贫风险评价指标体系,使用集成学习的多种算法构建返贫风险预警模型。结果表明:各模型经过参数优化后,深度森林算法识别效果最佳,对返贫风险识别整体准确率为0.98,对构建动态返贫风险预警系统有一定的技术帮助;各地区不同风险等级生计资本存在显著差异性,应因地制宜出台返贫帮扶政策,并为精准帮扶防范返贫提出相应建议。
In order to ensure further consolidation of the effectiveness of poverty eradication,coordinate and expand the articulation of poverty eradication results and rural revitalization,and lay a good foundation for the implementation of the rural revitalization strategy,the 2020 China Rural Revitalization Survey(CRRS)data was selected as a sample,and a return-to-poverty risk evaluation index system was constructed under the guidance of the theory of sustainable livelihoods,and the return-to-poverty risk early warning model was constructed using multiple algorithms of integrated learning.The results show that after parameter optimisation,the deep forest algorithm is the most effective in identification,with an overall accuracy of 0.98 in identifying the risk of returning to poverty,which provides some technical help for the construction of a dynamic return to poverty risk early warning system.There are significant differences in the livelihood capitals of different risk levels in different regions,and poverty alleviation policies should be introduced according to local conditions,so as to put forward corresponding suggestions for precise assistance to prevent poverty returning.
作者
郭文强
谭乔阳
雷明
马志龙
GUO Wenqiang;TAN Qiaoyang;LEI Ming;MA Zhilong(School of Information Management,Xinjiang University of Finance and Economics,Urumqi Xinjiang 830012,China;Guanghua School of Management,Peking University,Beijing 100871,China;United Front Work Department,Party Committee of Xinjiang University of Finance and Economics,Urumqi Xinjiang 830012,China)
出处
《长沙大学学报》
2024年第2期1-8,共8页
Journal of Changsha University
基金
国家社科基金西部项目“边疆民族地区规模性返贫的防范与治理机制研究”,编号:23XMZ060
新疆社科基金一般项目“新疆脱贫攻坚与乡村振兴有效衔接模式及实现路径”,编号:21BGL098
新疆财经大学研究生科研创新项目“基于集成学习算法的返贫风险预警及防范研究”,编号:XJUFE2023K033。
关键词
返贫风险
生计资本
深度森林
risk of return to poverty
livelihood capital
deep forest
