Unveiling the Connection: Music Preferences and Mental Health Through Machine Learning Analysis

A song can find its way into your heart, and then transform the way you think, as famously conveyed by Elvis Costello. Our emotions shape the type of music we are listening to, and our music preferences impact our emotions as well. During rainy days, we feel sadness gravitating us toward slower music echoing the sentiments. To find inspiration, we listen to pop and rock to boost us up. Music therapy brings benefits to people of all ages, regardless of their musical expertise or health. Music can change our mood, alter body rhythms, and divert our focus, affecting our behavior. The objective of this research is to design a model that determines the possible state of a person's mind based on their usual music preferences. The data we used focused on the effects of music primarily on depression, anxiety, OCD, and insomnia. During our research, we compared 5 different models and saw that Random Forest and K-Nearest Neighbor gave the best results with accuracy reaching up to 85%. Our exploratory data analysis identified that individuals reporting high levels of depression, OCD, and insomnia prefer Lofi music, while those with higher anxiety levels tend toward Folk and K-pop. In conclusion, specific musical preferences can be linked to different conditions, such as anxiety and depression. The model created in this research enables identifying the effect of music on mental health with good precision. Eventually, this process of incorporating music into mental health management can be personalized.