Analysis And Research On Self-Learning Models Based On Learning Behavior Sequences

With the continuous development of engineering education,improving students’ innovative ability has gradually become the future development direction,and the improvement of innovative ability needs to be based on the ability of self-learning.Therefore,in engineering education,more and more attention is paid to the cultivation of students’self-learning ability.However,the existing educational theories cannot fully play a guiding role in self-learning scenarios,which further makes it difficult to carry out quantitative assessment of students’ self-learning ability and mining and analysis of self-learning models.This thesis focuses on the sequence of students’ learning behavior,and from the perspective of learning theory,proposes a learning theoretical model suitable for self-learning scenarios,excavates the self-learning mode in the sequence of students’ learning behavior,and quantitatively evaluates the students’ self-learning ability.The research contents of this thesis are as follows:(1)Aiming at the limitations of data analysis in self-learning scenarios based on the "zone of proximal development" and "scaffolding" theories.Combined with the analysis of past teaching data,we found the limitations of "zone of proximal development theory" and "scaffolding theory" in autonomous learning scenarios,and constructed theoretical models based on autonomous learning,namely "zone of meaningful development theory" and "generative self-scaffolding theory".From the theoretical level,explain the conceptual details of the proposed model in detail to verify the correctness of the theoretical model;from the application level,design corresponding teaching cases to prove the feasibility of the theoretical model;(2)Aiming at the lack of objective quantitative evaluation methods for autonomous learning ability.Combined with learning sequence data and dynamic evaluation theory,a self-learning ability evaluation algorithm based on the new learning theory model is designed.Use the learning behavior sequence data to evaluate the students’ self-learning ability,and do correlation analysis and multiple linear regression analysis with the students’ academic performance to verify the correctness of the evaluation algorithm and the effectiveness of the learning theoretical model;(3)Aiming at the lack of research and complete analytical framework for the self-learning model.According to the constructed learning behavior sequence and learning theoretical model,an analysis framework of self-learning mode based on SGT algorithm is proposed.Through sequence feature extraction,sequence clustering and lag sequence analysis,the different learning modes of students in self-learning scenarios are excavated,and training suggestions suitable for self-learning modes are put forward in combination with students’achievements.