基于数据场K-means聚类的农村贫困人口精准分级研究——以贵州省某镇为例  

Classification of Rural Poor Population Based on Data Field K-means Clustering——A Case Study of a Town of Guizhou Province

作　　者：龚艳冰 巢妍[1] GONG Yan-bing;CHAO Yan(School of Business Management , Hohai University, Changzhou 213022;Institute of Statistics and Data Science, Hohai University, Changzhou 213022)

机构地区：[1]河海大学企业管理学院,江苏常州213022 [2]河海大学统计与数据科学研究所,江苏常州213022

出　　处：《软科学》2019年第6期135-139,共5页Soft Science

基　　金：教育部人文社会科学研究青年基金项目(18YJCZH036);江苏省高校优秀中青年教师和校长境外研修计划项目(2018)

摘　　要：综合考虑贫困人口分级单指标和多指标,提出了一种基于数据场K-means融合聚类的农村贫困人口精准分级方法,该方法先由数据场势函数得到初始聚类的个数与聚类中心,再将其导入K-means聚类算法得到最终分级结果,有效地解决了传统K-means算法需要主观给定聚类参数的问题。最后,以贵州省某乡镇贫困人口数据为例进行实证分析,结果表明,该融合聚类方法更简洁、高效,能够为农村贫困人口分级提供科学合理的参考。Considering the single index classification and the multi-index classification of the rural poor, an accurate classify method for the rural poor based on data field and k-means clustering is proposed which combines the data field with the k-means algorithm. The method firstly obtains the number of clusters and the cluster center determined by the potential function. Then, it uses the K-means clustering algorithm to obtain the final grading result, which effectively solves the problem that the traditional K-means algorithm needs subjectively given parameters. Finally, an empirical analysis is made, taking a town in Guizhou Province as an example. Result shows that the method is effective and feasible, providing a new idea for the rural poor classification.

关 键 词：精准扶贫 数据场 K-MEANS聚类 分级 

分 类 号：F323.8[经济管理—产业经济]