Author Identification from Song Lyrics

Abstract

Machine Learning (ML) tools have been used extensively in a wide variety of domains recently. Due the enormous amount of data being produced, machine learning techniques are being heavily used to make sense of data & derive meaningful results. Using machine learning tools, we can turn the data into knowledge. Music is one of the truest forms of art. Bangladesh has a great history of music with a great tradition of song writing over centuries. Authorship attribution is the way of identifying the author from a linguistic corpus. This paper demonstrates a guideline to identify the author of a Bengali song from the lyrics of that song using machine learning. This research work presents the first work on machine learning approach for author attribution from the lyrics of a song. Here six methods of machine learning are used for the author identification and high accuracies have been achieved from these methods. It is observed that Naïve Bayes method provides higher accuracy in comparison with the other methods.