Collaborative Recommendation for Scenic Spots based on Degree Centrality
Jinlong Chen, Yiming Jiang, Minghao Yang · 2019
In the collaborative filtering recommendation algorithm, similarity calculation shows a significant influence in recommended quality. Only by looking at the scores of all users and aiming to calculate the similarity of the normal recommendation algorithm, it is impossible to accurately reflect the user demand for the product. In order to eliminate these interferences, this research puts forward an algorithm based on degree centrality. The algorithm introduces the influence of the user in the similarity calculation, judges the influence according to the degree Centrality of the user's degree, and calculates the user's rating records on the target item for recommendation. Experimental validation: Compared to the original user-based recommendation method, the proposed recommendation algorithm based on degree centrality of this paper has certain improvements in Mean Absolute Error, Root Mean Square Error, F-rneasure, Coverage and Precision.