Visualization of Music Plagiarism: Analysis and Evaluation

Nowadays plagiarism is an interesting and debated topic in different fields. In music, the plagiarism is a very common phenomenon which touch the vast amounts of money that music melodies are able to generate in today's pop music market. In a music composition, the melody is assumed to be the most significant factor in a court's decision about whether a new music composition is an illegitimate version of a pre-existing composition. Despite the wide-spread belief that there is a fixed and trivial number of corresponding notes between two melodies, the similarity analysis is a very complex process. In this paper we address the plagiarism in pop music, and specifically, we study whether visualization can facilitate the task of discovering melodic similarities among musical songs. To investigate this, we defined three representations to show the melodic relations among songs. We performed a user study in which subjects performed different tasks on a song collection using these representations to investigate which one is best in terms of intuitiveness and accuracy. Results of the study provided us with positive feedback as well as further directions to explore.