Pemanfaatan Matrix Factorization Berbasis Android Berdasarkan Preferensi Pengguna untuk Optimasi Promosi Objek Wisata

This research aims to optimize the promotion of tourist attractions in Simalungun Regency and Pematangsiantar City by utilizing Artificial Intelligence (AI) technology based on Matrix Factorization. The lack of information about tourist spots in these areas has resulted in many attractions being relatively unknown to the public, making current promotional efforts less effective. Through the application of AI, it is hoped that the promotion of tourist attractions will become more efficient, providing users with a better experience in finding destinations that match their preferences. The data used in this study was collected through web scraping from Google Maps, including information about tourist spots such as name, location, rating, number of reviews, and category. This data was then processed using Matrix Factorization, which analyzes user preferences based on their visit history and ratings of tourist attractions. The ultimate goal is to generate personalized and relevant recommendations. The research was implemented in the form of an Android application, allowing users to easily access tourism recommendations via mobile devices. Testing results showed that the recommendation system successfully provided suggestions for tourist attractions that matched users' needs. The model was evaluated using Root Mean Squared Error (RMSE), yielding a result of 0.0835 on training data and 0.2362 on validation data, demonstrating good performance. Overall, this research proves that Collaborative Filtering technology based on Matrix Factorization is effective for tourism recommendation systems.