Research On Learning Resources Recommendation Based On Online Learner Test Results And Comment Data

With the development of information technology,online learning has gradually become an important way for people to acquire knowledge.However,with the expansion of the online learning community and the increase in the number of learning resources,it is difficult for learners to enter the online learning community to learn.It takes a lot of time for learners to find learning resources that suit their learning style and learning needs.This can lead to inefficient learning and loss of interest in learning.In order to improve the learning efficiency of the learner,the learner can quickly obtain the required online learning resources in the learning process,and analyze the online test scores and comment data generated by the learner to form a learning resource recommendation list for the learner.The research content of this paper is carried out from the following aspects:First,collect online data generated by learners and build a learner model.Among them,online data mainly includes learner online test score data and online comment data.By analyzing the learner's online test score data,the learner's mastery of each knowledge point is obtained,and the learner's mastery of the learning resources is constructed.At the same time,the natural language processing technology is used to mine the review data generated by the learner,and the learner's mastery of the learning resources is obtained,and then the value of the corresponding item in the learner-learning resource mastery matrix is corrected.The implicit semantic model is used to complete the decomposition of the matrix,and the characteristics of the learner and the learning resources learned by the learner are obtained.Secondly,the training process of the convolutional neural network is completed by using the learning resource features and corresponding text description information obtained by the implicit semantic model,and the learning resource model is constructed.The text description information of the unknown learning resource is processed through the trained learning resource model to obtain the learning resource feature.Third,a learning resource recommendation list is generated for the target learner according to the learner characteristics and the learning resource characteristics.The similarity between the target learner and the learning resource is calculated by using the acquaintance calculation method of the vector,and the learning resource whose similarity value is greater than the threshold is selected and added to the learning resource list.At the same time,the collaborative filtering recommendation algorithm is used according to the similarity between the learners,and the learning resources that other learners pay attention to during the learning process are recommended for the target learner.In summary,in order to reduce the time for learners to acquire learning resources in the online learning process and improve the learner's learning efficiency,the learner is obtained by analyzing the test score data and the review data generated by the learner in the actual course learning process.And the characteristics of the learning resources,the learner's personalized learning resource recommendation,verify the effectiveness of the algorithm.