Application of Hybrid Filtering Strategies in Music Recommendation System

Abstract

Everyone has their own distinct musical preferences; it's safe to assume that each music will find an appreciative audience. It's important to note that there isn't a single human society that has ever survived without music. There are two major gains from this study. Initially, a multi-strategy approach is taken to develop hybrid recommendation algorithms that give more accuracy than the existing algorithms. Also this hybrid algorithm is used to find new music in real time. This allows the algorithm to make an educated guess as to which musician and song best suit the user. As a second step, a general context-aware and emotion-based customized music framework is offered to facilitate the quick growth of context-aware music recommendation systems and to shed light on the whole recommendation procedure. Multiple methods exist for responding to requests, and a general framework is required for both collecting these methods and interpreting them within the context of the proposed framework. The kind of recommendation algorithm used is decided by the format of the input.