Research on Collaborative Filtering Recommendation Algorithm Optimization in Study Tour Route Recommendation System

This paper addresses two primary challenges in personalized Tour Route recommendation systems using collaborative filtering algorithms: data sparsity and scalability. To resolve data sparsity, it introduces the Slope One algorithm for filling missing values in the initial rating matrix and then applies a K-means clustering-based collaborative filtering algorithm for rating predictions. For scalability issues, the paper proposes an enhanced K-means algorithm with centroid agglomeration parameters. Comparative experiments using the MovieLens dataset show that these improvements significantly enhance recommendation accuracy, effectively addressing the problems of data sparsity and scalability. The study's findings offer valuable insights into collaborative filtering recommendation algorithms and provide a theoretical framework for refining the accuracy of recommendation systems.