Multiple Techniques for Raga Identification in Indian Classical Music

Data mining can be performed on Indian classical music with the help of raga. Raga plays a vital role in Indian classical music. Raga consists of different characteristics which make every raga unique in it. Raga is sequential collection of swaras and it is divided into two systems Hindustani (North Indian) music, Carnatic (South Indian) music. In this work we discussed basic introduction about data mining and raga identification and investigated characteristics of raga, manual method for raga identification. We survey the different past raga identification techniques such as scale matching, arohanaavarohana pattern, statistical approach and Pakad matching, Pitch class distribution (PCD) and pitch class dyad distribution (PCDD), An unsupervised statistical approach: LDA model, swara intonation. We introduced these techniques with their system structure and results obtain by them. We identified the challenges associated with raga identification, and related future work.