摘要
目前高校贫困生的精准认定是一个公开的难题。本文以某高校连续60天的校园卡消费记录为依据,利用Python语言和K-Means聚类算法,依据15545名大学生个人消费金额,而将他们分5个"最优群体"。从最低消费群体中挖掘应该被认定为贫困生而没有被认定为贫困生的群体,从最高消费群体中挖掘不应该被认定为贫困生而被认定为贫困生的群体。本文以客观的消费记录为标准,利用大数据挖掘技术,为科学资助和精准资助提供了决策支持。
This article bases itself on the campus card consumption of 15545 students from a certain university for 35 consecutive days.It uses Python language and K-Means clustering algorithm to categorize these students into 5optimal groups according to the total amount of their personal consumption,picking out the ones from the group that consumed least who should have been identified as needy and the ones from the group that consumed most who should not have been identified as needy.The paper attempts to provide some decision support for financially aiding students in a scientific and precise way.
出处
《高校辅导员学刊》
2016年第5期74-77,共4页
Journal of College Advisor
关键词
大学生
贫困生
K-MEANS
分类
判定
college student
needy student
K-Means
classification
identification
