LAUSR

The accuracy of a recommendation is an important index to evaluate the performance of a recommendation system. The personalized recommendation system tends to pay too much attention to the accuracy of recommendation results and often neglects the diversity of the recommendation results. In this paper, domain ontology is used to construct the user interest model, and the integrated ontology-based semantic similarity algorithm is used to obtain the user ontology set. Then, the semantic interest community is constructed through the hierarchical clustering method. Users with a high degree of diversity are selected as trusted neighbors to construct a hybrid recommendation model with a combination of accuracy and diversity. The experimental results show that the hybrid model can improve the diversity of the recommendation system by adjusting the weight factor while having less influence on the accuracy.