Apriori Algorithm-Based Learning Behavior Mining for Mobile Education Platforms

In order to promote learners' learning effectiveness and improve the accuracy of learning behavior mining, this paper conducts research on the Apriori algorithm based learning behavior mining on mobile education platforms. Firstly, web crawler technology is used to capture the behavioral information of learners during the learning process on the mobile education platform to construct learner profiles, and preprocess the sub-network set data of learning behaviors. Secondly, a Hash table is constructed to improve the Apriori algorithm to extract the learning behavior characteristics of learners on the mobile education platform. Then, a Stacking ensemble learning model is built to determine four base learners for model training. Finally, the Stacking ensemble learning model is improved with a chain rule, and conducting k-fold cross-validation to achieve data mining of learning behaviors on the mobile education platform. Comparative experiments have proven that when using the method proposed in this paper for data mining of learning behaviors on the mobile education platform, the normalized difference precision is always above 90%, the mAP value is always above 93%, the mining scope coverage rate is maintained above 90%, and the comprehensiveness is kept above 90%. This indicates that applying the method proposed in this paper to the data mining of learning behaviors on the mobile education platform can improve the accuracy of data mining and has a good mining effect.