Songs Popularity Analysis Using Machine Learning

It is generally agreed that music is the common tongue of humanity. Nowadays it is hard to imagine our life without music. It has a big influence on our daily life. Over the centuries, people appreciated music in their lives. Listening satisfaction, emotional reaction, performance, and creation were all important to them. This is true in both classical and contemporary concert music. For others, music is seen as an escape, an oasis that offers a chance to escape from the hardships of life. It provides people with relief and helps with stress reduction. Music is used as a powerful therapy to calm down or to cheer up in the moment of joy. Music is not a simple source of amusement, it also develops the mind and helps with self-confidence. The algorithms of music recording are not perfect. The sound loss happens while transferring analog signals to the digital. With technological development, it is possible to reduce the gap. Analog sounds usually approximated to become digital. Marching learning approaches help with the mathematical analysis of songs. Its algorithms allow predicting songs' popularity. The influence of different characteristics is studied on the example of Korean music. The study explains the uniqueness and popularity of BTS Korean band music. The unlimited potential for technology growth, combined with machine learning, will take music to the next level. In order to achieve a high response from music fans, the artists should also take into consideration the analytics from machine learning algorithms. Its results based on a lot of songs, created over the centuries. It is a big and reliable source to increase the popularity of artists and improve their creation. The parameters considered in the master's thesis will be an excellent start for improving music in the future. Moreover, knowing what song will become a hit would help the music industry to raise their profit, as well as success. Songs' popularity predictions will also have a profound effect on businesses that depend on the music. Keywords: Music, Popularity, Machine Learning, Approximation, Digital signals, Analog signal, BTS