Collaborative Filtering Algorithm-Based Destination Recommendation and Marketing Model for Tourism Scenic Spots

Abstract

The information age of rapid development of tourism industry provides abundant travel information, but it also comes with the problem of information overload, which makes it difficult to meet the growing personalized needs of people. The traditional collaborative filtering recommendation algorithm (CFA) also suffers from the problem of data sparsity when the user population increases. Therefore, this study optimizes the CFA through the similarity factor and correlation factor and enhances the tourism sense of travel experience through the satisfaction balance strategy. The experimental results show that the improved CFA method has the highest average accuracy on the overall dataset and the best recommendation performance of the satisfaction balance strategy. Overall, the recommendation model in this study is useful for attraction selection of users and marketing optimization of travel companies.