Research on Education Resource Recommendation Algorithm Based on Implicit Feedback and Pairwise

In view of the problem of how to improve the accuracy of education resources recommendation in Internet era, a recommendation algorithm based on implicit feedback and pairwise education resources are proposed. Firstly, the feature extraction of learner-education resource is carried out, and the implicit features of learner-education resource are extracted by implicit semantic model, and the prediction results of factor decomposition are used as input to sorting learning algorithm. Converting sorting problems to classification problems uses the theory of ordering learning. Based on the experiment comparison with the traditional recommendation algorithm, it is compared with the comparison algorithm, which has better effect on the recommendation of educational resources. The algorithm increases by nearly 30% on metrics MAP, NDGG.