Learning effect optimization method based on user behavior causal relationship in MOOC log data

Abstract

The invention discloses a learning effect optimization method based on a user behavior causal relationship in MOOC log data. The method comprises the steps of based on log data of MOOC platform users,by means of user tendency calculation and matching, selecting and generating causal independent variables in a causal network according to a causal reasoning framework of a reverse fact, and obtaining the minimum data scale under different data scales according to the average Markov blanket length of the causal network when the length tends to be stable; obtaining a causal network group; screening edges among the network nodes by using an expert accuracy algorithm; grouping the screened network groups; comprehensively generating a final causal network by using a Bagging voting mechanism; obtaining the causal relationship between the user behavior and the learning effect; based on the causal relationship, reasonably planning the learning path of the user according to the reason variable node and the result variable node, the probability that whether the user completes the course is influenced by changing the operation behavior of the user or the learning time of the user, and the accuracy of relationship judgment between the variables is improved.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence, and in particular relates to a learning effect optimization method based on user behavior causality in MOOC log data. Background technique [0002] MOOC (Massive Open Online Course, MOOC) is a large-scale open online learning platform. The emergence of MOOC has broken the unbalanced distribution of educational resources, realized the networking of educational resources through the Internet, and made education in universities more open and Sharing and timeliness enable the whole people to have access to high-quality learning resources. Since the MOOC learning platform contains a large number of online courses and the massive data generated during the learning process of learning users, this makes it possible to observe the learning mode and path of each learning user in a timely manner while maintaining the scale, and optimize the learning process for users. , improve the quality ...

AI Technical Summary

Problems solved by technology

We still lack favorable methodological tools for the study of lack of symmetry

At the same time, most studies only care about the predicted results and ignore causal factors, which leads to the exploration of the experimental results after intervention only relying on correlation exploration, which greatly reduces the authenticity of the conclusions

Images