基于决策树算法的研究生遴选质量评价

本文针对如何提高研究生遴选质量、选拔出更多优秀生源的问题,提出一种基于决策树算法的研究生遴选质量评价方法。首先通过分析研究生生源学校以及初试和复试等招生信息,同时结合对研究生的课程学习成绩、参与科研项目情况、硕士毕业论文质量的跟踪,建立了适合于计算机专业研究生质量的评价指标。然后采用经典的ID3决策树算法对相关数据进行分析挖掘,以评价现有研究生招生体系中各项指标对研究生培养质量的影响,并通过统计学方法对结论进行逆向分析验证。结果表明在研究生入学考核的各项指标中,面试成绩和上机考试成绩在区分考生能力、优秀研究生遴选中具有关键作用。

Quality evaluation of graduate selection based on decision tree algorithm

Aiming at the problem how to improve the quality of graduate student selection in computer science major and select excellent students,a method of graduate student selection quality evaluation based on decision tree algorithm is proposed.First of all,by analyzing the source school of graduate students and the enrollment information,such as the primary and secondary examination,combined with the tracking of the graduate students’academic performance,participation in scientific research projects,and the quality of the master’s thesis,an evaluation index suitable for the quality of graduate students majoring in computer science is established.Then,the classic ID3 decision tree algorithm is used to analyze and mine the relevant data to evaluate the impact of various indicators in the existing graduate student enrollment system on the quality of graduate student education,and the conclusion is verified by reverse analysis through statistical methods.The results show that the interview scores and computer test scores play a key role in distinguishing the ability of examinees and selecting excellent graduate students.