Research on Scenic Spots Recommendation Algorithm Based on Tourism Big Data

Abstract

In order to solve the data sparsity problem existing in the traditional collaborative filtering recommendation algorithm, a collaborative filtering recommendation algorithm for scenic spots based on multi-dimensional feature clustering was proposed. Firstly, the users are clustered and classified according to the feature vector. Then we determine the category of the target user. Building user-scenic spot score matrix, on this basis, the user-scenic spot attention matrix is added. In order to optimize the traditional similarity recommendation algorithm, the attention matrix and the score matrix are linearly combined with the balance factor to calculate the similarity between users. In addition, the similarity threshold is introduced to determine the similar neighbor set. And recommend scenic spots to the target user according to the users in similar neighbor set. Finally, the MAE of the algorithm and the traditional recommendation algorithm are compared by using the tourism related data of Qingdao City crawled on the Mafengwo tourism website. The experimental results show that the algorithm proposed in the paper not only reduces the sparsity of data, but also improves the recommendation accuracy and has better stability.