Font Size: a A A

Study On The Auxiliary Discriminant Of Poor Students Based On Campus Card Data Mining

Posted on:2019-02-10Degree:MasterType:Thesis
Country:ChinaCandidate:B H ZhaoFull Text:PDF
GTID:2428330623968774Subject:Engineering
Abstract/Summary:PDF Full Text Request
Campus card has been widely used in many colleges and universities,and students produce a lot of data in the process of consumption,which hides a lot of information.It has become an important research direction for the information mining of campus card data.Data mining is the discovery of hidden information from massive amounts of data.The commonly used data mining methods include cluster analysis,classification,association analysis,etc.This paper focuses on the k-means algorithm in clustering algorithm.For the algorithm to be sensitive to the initial clustering center and the weight of each attribute is the same,a weighted Euclidean distance algorithm based on average density is proposed,and the improved algorithm is applied to the clustering of campus card consumption data.The experimental results show that the improved algorithm has higher clustering precision and the clustering results are more stable.The judgment of poor students in colleges and universities is a difficult problem that has long puzzled the management of colleges and universities.Since the distribution of financial aid is related to the vital interests of poor students,it is particularly important to correctly identify poor students.It is very necessary to excavate the hidden information in consumption data to make scientific basis for poor students.This paper will focus on the students who have normal consumption in the school canteen(that is,they reach a certain amount of consumption in that month),and the students with good habits of consumption and excellent academic performance are the key recipients.Due to the different data sources and large amount of data,it is necessary to preprocess the data.After preprocessing the card data of campus,this paper excavates and analyzes it.Firstly,the consumption data of a card is analyzed by cluster analysis.According to the clustering results,the consumption level of students is divided into three grades.Then according to the student for the first time this judgment standard charge time,analysis the consumption habit of the students,the consumption habit of the students can be divided into regular and irregular(reaches a certain number of times that rule during the month,or irregular).In the end,it is related to the student's academic performance.The data related to consumption level,consumption habit and academic performance are obtained,and the standard of auxiliary judgment of poor students is obtained,and then the scientific decision-making Suggestions are put forward for managers.
Keywords/Search Tags:Campus card, Cluster analysis, k-means, Consumption, Auxiliary judgment
PDF Full Text Request
Related items