Research On Course Resource Recommendation Technology Based On Deep Learning

While we are now in the era that the Internet is booming,traditional off-line teaching method has been unable to meet the urgent need of users for learning resources.Therefore,choosing on-line education has become a new way for more and more learners,because this method provide a new way that is not restricted by time and space.With the continuous development of online education,we often meet some trouble choosing education resources when we acquire resources conveniently.The ever-increasing number of online course resources is dazzling and incomprehensible,It may take a long time for learners to find and accurately find the effective learning resources they need.Therefore,how to quickly and accurately find the resources that users need from the kinds of resources has become an urgent problem to be solved,and the recommendation of education resources can solve such problems effectively.In recent years,the technology of recommendation has been widely used in various fields and has made considerable progress.My paper will apply the recommendation algorithm to the online education environment,analyze the learning records of learners to learn user preferences,and provide users with corresponding courses according to these needs.By integrating deep learning into traditional recommendation algorithms,and using deep learning's powerful ability to discover potential connections,it can not only alleviate the problem of data sparseness,but also improve the accuracy of recommendation.This paper mainly includes the following contents:(1)Introduced the basic situation of online education and the current research status of curriculum resource recommendation.The research background and significance of recommender system are analyzed,and the recommender system and some common recommendation algorithms are briefly explained.(2)According to the analysis of the current research status and development of the related technology of curriculum resource recommendation,and the problem of sparse data and low accuracy in current curriculum recommendation,deep learning technology is applied to curriculum resource recommendation.Aiming at the problem that the importance of learning resources to users will change over time,this paper proposes to integrate time information into a neural collaborative filtering algorithm through a clustering classification algorithm,and proposes a course resource recommendation algorithm based on deep learning(improved Neu MF algorithm),To better recommend the courses that users want to learn at this stage.(3)In order to verify that the course resource recommendation algorithm based on deep learning has good recommendation ability under both explicit feedback and implicit feedback.This paper crawls the real data on the MOOC website as a data set,selects MAE and RMSE as evaluation indicators under explicit feedback,and selects NDCGand HR as evaluation indicators under implicit feedback.And compare with CF algorithm,BP neural network,NCF algorithm.Experiments have proved that the model proposed in this paper has a good recommendation effect no matter it is under explicit feedback or implicit feedback.(4)In order to better understand the application of course recommendation in the course resource learning platform,this paper carries out the design of the electric course resource system based on the recommendation system to simulate the application of course recommendation in reality.