Personalized knowledge tracking method and system fusing learning behavior characteristics

Abstract

The invention discloses a personalized knowledge tracking method and system fusing learning behavior characteristics. The method comprises the following steps that effective characteristics in a composite vector composed of learning behaviors and answer result data by using a convolutional neural network are extracted; question information features including knowledge points are extracted through a noise reduction auto-encoder, finally, the learning behavior features and the question information features are combined, and the knowledge mastering degree state of students is obtained through an LSTM network and a full connection layer. In the modeling process, a series of learning behavior characteristics of the students in the learning process and rich information of exercises and knowledge points are fused, and the knowledge mastering degree of each student is predicted more accurately. The method can be applied to hybrid teaching and provides a quantitative basis for personalized teaching.

Description

technical field [0001] The invention relates to the technical field of blended teaching, in particular to a personalized knowledge tracking method and system that integrates learning behavior characteristics. Background technique [0002] Blended teaching is the deep integration of online and offline teaching modes. At present, the rapid development of information technology makes the blended teaching mode more and more applied to teaching. Blended teaching not only retains the opportunity for teachers and students to communicate face to face, but also breaks through the time and space constraints to the greatest extent. Learners can use online platforms Teaching activities such as pre-class preview, flipped classroom, and after-class review and consolidation, these teaching activities allow students to leave traces on the teaching platform and generate learning behavior data. Usually, the teaching platform will provide visual windows, logs and other forms to display the st...

AI Technical Summary

Problems solved by technology

Although the technical improvements of these models have achieved good results, most studies only consider knowledge points and answer results, ignoring the impact of learning behavior performance on knowledge mastery; Behavioral performance, such as the time spent on answering questions and whether to check the analysis, etc., so the prediction accuracy of knowledge tracking needs to be improved

The performance of learning behaviors was not included in the research. The author believes that one of the main reasons is that before the general development of educational platforms, it was difficult to obtain data during the learning process of students, which made it difficult to quantify the impact of learning status on learning outcomes, thus limiting the advancement of research

Images