Automatic music classification for Dangdut and campursari using Naïve Bayes

Music classification can be performed by classifying music according to its genre, style, mood, and others. Various methods have been implemented to automatically classify music. Naïve Bayes learning algorithm is one of the most efficient and effective classification algorithm. Dangdut and campursari music are the music often heard by Indonesian. But the classification of dangdut and campursari music is still rarely performed. In this study, we perform automatic music classification for dangdut and campursari music. We use Naïve Bayes to classify music and the data was discretized based on Minimum Description Length Principle (MDLP). We used jSymbolic to extract feature from MIDI files. Currently, we use 45 features that are included in the category of instruments and pitch. This experiment produced the accuracy of 85.14%.

I. Introduction

Currently, music lovers can enjoy music from various types of digital media, such as MIDI, MP3, and WAV. Typically, digital music collections contain a large number of relevant digital documents for one musical work, which are given in various digital formats. Since, the manual of descriptive labels is infeasible for large datasets, one needs fully automated procedures for data annotation as well as efficient content-based retrieval methods that only access the raw data itself without relying on the availability of annotations [1].