A Preferred Music Recommendation Method for Tinnitus Personalized Treatment Based on Signal Processing and Random Forest

摘要

To relieve tinnitus, a new scheme used non-repetitive preferred music was proposed and proved to be effective in short term. However, tinnitus treatment is a long-term process. Whether we can provide preferred music to tinnitus patients accurately and efficiently is the key to long-term implementation of this new scheme. Therefore, this paper proposed a dedicated recommendation method for tinnitus personalized treatment based on signal processing methods and random forest. Through it, we can predict preferred degree of unknown music, then recommend and develop preferred music for tinnitus patients accurately and efficiently. It might be helpful for long-term implementation of the new scheme.