基于人工智能LSTM循环神经网络的学习成绩预测

The Evaluation and Prediction of Academic Performance Based on Artificial Intelligence and LSTM

学习成绩预测能够助力学生课业学习、提升教师教学能力、协助学校评估教学质量和优化教学管理,已成为教育领域研究的热点与难点。文章以学生各阶段历史成绩为基础,结合考勤、宿舍卫生、校园纪律等行为特征数据,利用人工智能LSTM循环神经网络模型对课程成绩进行预测。基于预测课程的成绩,可以对存在潜在挂科风险的学生提出学业警示;对教师改进教学方法、优选教学手段、优化教学过程、提高教学质量提供帮助;同时协助学校开展教学管理进而预防教学事故的发生。实验结果表明,该方法能较准确地预测学生的课程成绩,具有一定的有效性和实用性。

In order to serve students’learning,improve teachers’teaching ability and evaluate the quality of teaching,and optimize the teaching management,LSTM recurrent neural network model is used to predict students’course achievements based on the past achievements,attendance data,records of dormitory sanitation and violations of school rules of students.The experimental results show that the method can accurately predict the students’course performance with availability and practicability.The predicted results can warn the students who may fail the course,provide some suggestions for teachers to improve the teaching methods and means,assist the school to carry out teaching management,and prevent teaching accidents.This research has popularization and application value to a certain extent.