摘要
结合当前高校"精准帮扶"的现状,运用数据挖掘技术对受资助学生进行调研分析,选取8类贫困测评指标,尝试建立5级量化标准,建立贫困程度综合评估模型;采用主成分分析法,对贫困测评指标数据进行定量分析,利用特征值曲率谱最大峰值点方法有效保留受资助学生真实的贫困信息,减少通过观察或者经验模型带入冗余信息。此模型可以为大学生资助提供决策支持,对学生奖助学金发放、勤工俭学岗位设置,以及对新入库的学生进行贫困程度定级具有积极作用。
Combining with the current situation of "targeted poverty reduction"in colleges and universities,this paper analyzes the aided students by using data mining technology to select eight kinds of poverty assessment indicators,and tries to establish a level five quantization standard and a comprehensive assessment model of poverty level. This model adopts the principal component analysis method to quantitatively analyze the poverty evaluation index data,and proposes to select the effective eigenvalues by using the maximum peak point of eigenvalue curvature spectrum,reduces the redundant information by observing or experiential model. This model can provide decision support for college students' funding,and has a positive effect on student awards,work-study programs,and grading of students' poverty level.
作者
曹奇
CAO Qi(School of Creative Studies,Changzhou Vocational Institute of Textile and Garment,Changzhou 213164,China)
出处
《镇江高专学报》
2018年第3期82-86,共5页
Journal of Zhenjiang College
基金
2017年度常州纺织服装职业技术学院人文社科教育立项一般课题(RJ201716)
关键词
贫困程度
主成分分析
量化标准
曲率谱
精准帮扶
degree of poverty
principal component analysis
quantification criterion
curvature spectrum
targeted poverty reduction