ID3算法在程序设计类课程成绩分析中的应用研究

摘要

Based on the analysis of the main influence factors of studentsˊ performance, the ID3 algorithm is used to accomplish the decision tree. Algorithm consists of data acquisition, data processing, drawing decision tree, decision tree pruning, to noise and the accuracy of the model evaluation. Model selection of software engineering 1 class 2012 students of the "JAVA program design" course final exam results and studentsˊ basic information and other data as training data. After repeated training, the model can be based on the basic information of the students to analyze the studentsˊ learning performance, good, poor probability. The accuracy rate of the decision tree is 90%, which can satisfy the needs of users, and the decision tree model meets the requirements.%基于分析学生成绩主要影响因素的目的，采用了 ID3算法完成决策树的生成。算法包括数据采集、数据处理、绘制决策树、决策树剪枝、去噪和模型准确性评估等阶段。模型选取软件工程专业2012级1班的学生的《JAVA程序设计》课程期末考试成绩和学生的基本情况信息等数据作为训练数据。经过反复训练，模型能够根据输入学生的基本信息分析出学生的学习成绩优，良，差的概率。该决策树准确率为90%，能够满足用户需求，决策树模型符合要求。