基于ANN的大学生一般能力倾向对专业成绩的预测模型研究

Research on Predicting Model of University Students' General Aptitude for Major Achievement Based on ANN

应用人工神经网络模型探讨一般能力倾向对不同专业大学生专业成绩的预测作用。以652名大学生为研究对象,将一般能力倾向特征作为预测因子,利用Clementine数据挖掘软件分别构建文、理、工三类专业大学生一般能力倾向对专业成绩的预测模型。研究表明:三类专业大学生一般能力倾向对专业成绩预测的人工神经网络模型估计的准确率均在90%以上,平均绝对预测误差值在0.091到0.106之间。该人工神经网络模型对三类专业的专业成绩预测具有较高的准确性,一般能力倾向可以用来预测不同专业学生的专业成绩,为中学生升学填报专业提供依据。

It applied the Artificial Neural Network （ANN） model to predict the major achievement of university students in different majors with their general aptitude. 652 university students were selected as subjects. With general aptitude of the students in liberal, science and engineering as predictive factors, ANN model was built by using Clementine. The results indicated that the accuracy of ANN model＇s esti- mation towards the major achievement by analysis of general aptitude for university students in different majors is all above 90%. The average absolute error was between 0. 091 and 0. 106. The model with some precisions showed that the ANN model made an accurate forecast of the three majors＇ achievement. The general aptitude can be used to predict students＇ achievement of different majors and to provide references for high school students in choosing a major.