Context-aware Primary School Reading Assessment Considering Personalized Recommendation Algorithm

Abstract

With the development of the mobile Internet, reading information can be accessed by more and more means. However, how to obtain targeted information from the information has become a bottleneck in primary school reading. Hence, a mobile reading recommendation algorithm that combines contextual and collaborative filtering is put forward in this paper. Firstly, the naive Bayes method is applied to calculate the resource category with the highest preference for users in a specific context. Subsequently, the context similarity of the resources in the category is calculated to screen out the two-dimensional “user-resource” scoring model in the current context only or the context that is most similar to the current context. Further, the conventional user-based collaborative filtering algorithm is used to generate a recommendation list that meets the deeds of primary school reading. Experiments are carried out by capturing Sina blog posts in the background of primary school reading. The results suggest that the mean absolute error of the algorithm proposed in this paper is lower than that of the other related algorithms under the same conditions, and it has higher recommendation quality.