Recommendation of Tourist Attractions Based on Slope One Algorithm

With the development of domestic tourism, personalized demands for tourist attractions recommendation systems have increased. Since some drawbacks of traditional collaborative filtering include scalability and sensitivity to data sparseness, a new Slope One algorithm with attractions similarity is proposed in this paper. This Slope One algorithm weakens the effect of popular tourist attractions on prediction ratings, integrating attractions rating similarity and attractions attribute similarity calculation. Experimental results show that the proposed Slope One algorithm not only improves the accuracy of prediction effectively, but also optimizes the balanced distribution of recommendation results.