决策树和关联规则技术在高职院校招生工作中的应用研究

发布时间：2018-07-16 16:20

【摘要】：近年来,随着国家对职业教育的扶持力度不断加大,职业院校的规模迅速扩张,致使职业院校之间的竞争愈加激烈。在如此激烈的竞争中,各个职业院校要想占有一席之地就必须提高新生的报到率以及获得更多的优质生源。目前,一些院校建立了招生信息系统,希望借助于信息化的平台为招生工作提供帮助。但是在这些招生系统中,大部分仅是完成了信息统计和存储的功能,并不能解决实际的问题。而数据挖掘技术的出现,可以将存在于生源信息之间有价值的规律挖掘出来,找到影响学生报到率和生源质量优劣的因素,从而为学校的管理者科学的指导招生宣传工作、合理的制定招生政策提供帮助。本文从招生工作的实际出发,通过应用决策树和关联规则算法对学生的录取信息进行挖掘,分析研究影响学生报到率的因素。论文所完成的主要工作如下：1、研究探讨了数据挖掘的基础知识以及数据挖掘技术中的决策树和关联规则算法,并分别选取决策树算法中的ID3算法、C4.5算法和关联规则算法中的Apriori算法进一步研究分析,比较不同算法之间的差异。2、结合实际工作,对高职院校生源存档数据按照数据挖掘工作的要求进行数据预处理,保证后面的数据挖掘工作更高效、更可靠。3、对高职院校生源存档数据集分别运用C4.5算法和Apriori算法进行数据挖掘,并在WEKA平台上实现相关算法,得出相关规则和结论。通过对数据集使用上述两种不同的数据挖掘方法,可以分析研究存在于两种算法间的相关性、差异性,归纳出更符合实际需求的规则,从而为日后的招生工作提供帮助。
[Abstract]:In recent years, with the increasing support for vocational education, the scale of vocational colleges is expanding rapidly, resulting in more fierce competition between vocational colleges. In such a fierce competition, if vocational colleges want to have a place, they must improve the enrollment rate and obtain more high quality students. At present, some colleges and universities have set up enrollment information system, hoping to help recruit students with the help of information platform. But in these enrollment systems, most of them only complete the function of information statistics and storage, and can not solve the practical problems. The emergence of data mining technology can excavate the valuable rules existing in the information of students, find out the factors that affect the students' enrollment rate and the quality of students, so as to guide the enrollment propaganda work scientifically for the school managers. Make appropriate enrollment policy to help. Based on the practice of enrollment, this paper uses decision tree and association rules algorithm to mine students' admission information, and analyzes the factors that affect students' registration rate. The main work of this paper is as follows: 1. The basic knowledge of data mining, decision tree and association rule algorithm in data mining are studied. The ID3 algorithm C4.5 algorithm and the association rule algorithm Apriori algorithm in the decision tree algorithm are selected to further study and analysis, compare the differences between different algorithms. 2, combined with the actual work, The archived data of students in higher vocational colleges are preprocessed according to the requirements of data mining to ensure that the data mining work in the future is more efficient. More reliable. 3, using C4.5 algorithm and Apriori algorithm to mine the data set of higher vocational college students' archival data, and implementing the related algorithms on the WEKA platform, and draw the relevant rules and conclusions. By using the above two different data mining methods, we can analyze and study the correlation and difference between the two algorithms, sum up the rules that are more suitable to the actual demand, and provide help for the recruitment work in the future.
【学位授予单位】：郑州大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：G717.32;TP311.13

【相似文献】