Research on Intelligent Creation Based on Collaborative Filtering Algorithm

In the realm of intelligent music, composition research holds significant importance. However, it faces challenges with imprecise composition positioning. The prevalent bee colony algorithm fails to address this issue effectively in intelligent music composition studies, often yielding unsatisfactory outcomes. This paper introduces an innovative approach to intelligent music composition using a collaborative filtering algorithm and delves into its analysis. Initially, principles of proximity and similarity are employed to pinpoint influential factors, subsequently categorizing indicators to mitigate disruptive elements in composition research. Following this, the same principles aid in formulating a collaborative filtering algorithm-based strategy for composition research, leading to a thorough evaluation of results. MATLAB simulations reveal that when assessed against specific criteria, the collaborative filtering algorithm outperforms the conventional bee colony algorithm in both accuracy of compositional research and time efficiency related to influencing factors.