Hyperparameter Tuning On Machine Learning Transformers For Mood Classification In Indonesian Music

Music is also a form of art and expression of human feelings using regular or rhythmic sounds. In the last ten years, the development of digital music technology and streaming technology has resulted in wider access for everyone to enjoy music. Music evokes various feelings in the music lovers themselves. This is called the mood. In the field of computer science, music is an object that can be studied both through lyrics, audio, biographies and others. This research focuses on the mood of Indonesian music in the 70s and 80s by taking only the chorus. The data set used consists of text data and audio data with the transformer and bert models. The mood tested consisted of sad, happy and neutral. This area of interest is investigated using the Crisp-DM method derived from business understanding, data understanding, data preparation, modeling, evaluation, deployment. In constructing the Confusion Matrix, this study uses several tuning hyperparameters in the form of epoch, dimensions, set size, learning rate and others. The value generated by the confusion matrix contains a good value of 98%.