Application of Improved Collaborative Filtering Algorithm in Personalized Tourist Attractions Recommendation

With the advent of the era of information globalization, information in tourism industry is growing at an explosive rate in the Internet. More and more people hope to get high-quality tourism resources information more quickly and effectively. This trend is promoting the development of personalized tourism recommendation. In this context, the appropriate recommendation algorithm can make better use of tourism information and make the recommendation model have the ability to discover the potential value of users. However, tourist attractions are difficult to characterize because they contain many information such as geographical location, transportation route, cost information and user evaluation. Item-based Collaborative Filtering Algorithm (ItemCF) can skillfully avoid this situation, therefore we used ItemCF to build a personalized recommendation model for tourist attractions and apply it to real tourism dataset. The improvement of ItemCF is proposed in this paper, which can effectively improve the coverage of recommendation model. The biggest disadvantage of collaborative filtering algorithm is the cold start problem. This paper proposes a strategy to alleviate this problem. The experimental results show that the improved collaborative filtering algorithm has a good performance in the application of personalized tourist attractions recommendation.