A Personalized Recommendation Algorithm Considering Recent Changes in Users' Interests
Yiqi Gui, Xingchen Tian · 2018
As the present tag-based personalized recommendation algorithm does not consider the time factor, especially the short-term impact of recent interest on the recommendation results when constructing the user interest model, a collaborative filtering algorithm that combines user interest changes and tag features is proposed in this paper. The algorithm integrates the score information and the user's long-term and short-term interest factors into the calculation of label weights, and combines with the forgotten curve method to mine the user's real hobby. The experimental results show that the algorithm is run on the delicious-2k data set. The accuracy and interpretability of algorithm has been improved.