<![CDATA[The development of online music services demands increasingly personalized and contextual recommendation systems. However, most existing systems are still limited to processing historical data without taking into account the emotional state or activities of users. This study aims to design a machine learning-based music recommendation system that generates recommendations based on the user's listening history and artificial intelligence (AI) to produce personalized song recommendations based on the user's mood and activity. The methods used include song data analysis from the Spotify API, the K-Nearest Neighbor (KNN) algorithm, and the application of a Large Language Model (LLM) as a prompt-based interactive interface. Test results show that the system is capable of providing song recommendations with a 95% similarity rate using machine learning based on the songs listened to by users, and the application of AI produces more specific recommendations according to user prompts.]]>
Copyrights © 2026
