This paper investigates the application of exploring spatial clustering techniques in travel recommendation systems to improve recommendation accuracy and user satisfaction. A system that can provide personalized travel information has been designed by analyzing the spatial data and behavioral patterns of travelers. With this approach, the travel experience can be optimized by considering geographic location, user preferences, and tourist attraction characteristics. User data and attraction characteristics are analyzed in this paper to obtain accurate travel recommendations. Combining the traveler’s interest matrix and the attraction area heat algorithm is used to optimize the recommendation process. This recommender system reduces the root mean square error (RMSE) and mean absolute error (MAE) by 0.487% and 21-60%, respectively, compared with the traditional algorithm in the travel recommendation error analysis, and improves the accuracy rate to 96.33%, with a significant increase in the recall rate. The spatial clustering cluster analysis demonstrates that the chosen number of clusters can effectively enhance the clustering quality and maximize the recommendation results. By effectively utilizing spatial data of users and attractions, the system can provide personalized travel recommendations and improve user experience with high accuracy and recall, demonstrating better recommendation performance than traditional methods.