A Travel Recommendation Method Based on User Personalized Characteristics with Collaborative Fusion Matrix

With the improvement of people's living standards, people's spiritual needs continue to increase, which makes personalized travel recommendation more and more popular. However, the complicated and overloaded information in the internet have caused troubles for people to choose travel scenic spots. Although collaborative filtering is widely used for recommendation algorithms, neither user-based nor item-based collaborative filtering take the users' personalized characteristics into account. In order to recommend better travel scenic spots to users, we propose a personalized collaborative fusion algorithm combining user-based original collaborative filtering matrix with users' personalized preference matrix to make the recommendation more accurate and attractive. We build our experiments on the data of Qunaer, the results show that our proposed algorithms can not only recommend similar users' travel scenic spots, but also consider users' own personalized characteristics.