【摘要】 传统人工统计分析已无法满足大数据背景下全国计算机等级考试(NCRE)成绩分析。将数据挖掘中关联规则Apriori方法和决策树方法相结合,深入分析导致一级优秀率低及高级别通过率低原因。设计基于Apriori关联规则与决策树的成绩分析模型,建立题型与成绩、学生对课程的热爱程度等因素与成绩的关联。以2016年考试数据为训练集,2017年考试数据为测试集,根据实际成绩进行对比论证。实验表明,该方法可以有效帮助考生自主学习、发掘自身薄弱环节并针对性学习,同时帮助教师找准考试与知识点的核心,有针对性高效教学,使得一级优秀率平均提升25%,高级别通过率平均提升50%。更多还原

【Abstract】 Traditional manual statistical analysis cannot satisfy the analysis of national computer rank examination results under the background of big data. Using the combination of association rule Apriori method and decision tree method in data mining, this paper deeply analyzes the causes of low first-class excellent rate and low high-level pass rate. Based on the Apriori association rules and decision tree, this paper designs a performance analysis model, and establishes the relationship between test questions and scores,students’ love for the course and scores. Taking the 2016 test data as the training set and the 2017 test data as the test set, this paper compares and demonstrates them according to the actual results. Experiments show that this method can effectively help candidates to learn independently, discover their own weaknesses and learn pertinently. It can help teachers to find the core of exams and knowledge points, and teach effectively and pertinently. Using this method, the first-class excellence rate is increased by 25% on average, and the high-level pass rate is increased by 50% on average.更多还原