基于注意力机制的双向长短期记忆网络的在线工程实践评价框架
An Online Engineering Practice Evaluation Framework Based on BiLSTM with Attention Mechanism
马坤 1邵永伟 1郑楠 1陈贞翔 1杨波1
作者信息
- 1. 济南大学 信息科学与工程学院,山东 济南 250022
- 折叠
摘要
在虚拟学习环境下,工程实践课程通常以小组方式在线协同完成一个实践项目,但现有在线教学平台缺少基于学习行为数据的深度挖掘,教师难以像线下实践那样感知学习者的学习状态,从而无法进行十分客观公正的评价.为此,提出基于注意力机制的双向长短期记忆网络的在线工程实践评价框架,通过在线实践行为数据构建注意力机制的双向长短期记忆网络模型(GEP-BiLSTM)预测学生能否通过未来的实践考核.实践表明,所提方法可在学生出现学业预警前进行更有针对性的帮扶,从而提升工程实践的教学效果.
Abstract
In a virtual learning environment,engineering practice courses usually collaborate online in groups to complete a practical project.However,existing online teaching platforms lack deep mining based on learning behavior data,making it difficult for teachers to perceive the learning status of learners like in offline practice,thus making it difficult to conduct very objective and fair evaluations.To this end,a bidirec-tional long short-term memory network based on attention mechanism is proposed as an online engineering practice evaluation framework.Based on online practice behavior data,a bidirectional long short-term memory network model with attention mechanism(GEP-BiLSTM)is constructed to predict whether students can pass future practical exams.Practice has shown that the proposed method can provide more target-ed assistance to students before they encounter academic warnings,and can improve the teaching effectiveness of engineering practice.
关键词
教育数据挖掘/长短期记忆网络/实践评价/注意力机制/工程实践Key words
educational data mining/long short-term memory network/practical evaluation/attention mechanism/engineering practice引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据