PENDEKATAN PARTITIONING AROUND MEDOIDS (PAM) DALAM PENGELOMPOKAN GENRE MUSIK
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
Saragih, Parningotan
Kusuma, Dine Tiara
Karmila, Sely
Kusuma, Dine Tiara
Karmila, Sely
Subject
Teknik Informatika
Datestamp
2022-09-23 02:30:59
Abstract :
Music consists of many genres or subgenres, and it making difficult to set up a standard protocol system for standardizing or grouping a music genre if only use a subjective judgments. This research will discuss clustering music genre using PAM clustering technique (partitoning around medoids). The problem solving in this study will be using the "Muster" (Music Clustering) application, which is built using Matlab R2017 and making average energy (AE), tempo, time signature as parameter indicator and dividing as many as 5 clusters on PAM processing system built. After that it was tested three times experiment on Muster application, that is 10, 20 and 50 music data, and yielded accuracy level respectively 90%, 60% and 68%, so total accuracy of PAM in grouping music genre equal to 72.66%.
Music consists of many genres or subgenres, and it making difficult to set up a standard protocol system for standardizing or grouping a music genre if only use a subjective judgments. This research will discuss clustering music genre using PAM clustering technique (partitoning around medoids). The problem solving in this study will be using the "Muster" (Music Clustering) application, which is built using Matlab R2017 and making average energy (AE), tempo, time signature as parameter indicator and dividing as many as 5 clusters on PAM processing system built. After that it was tested three times experiment on Muster application, that is 10, 20 and 50 music data, and yielded accuracy level respectively 90%, 60% and 68%, so total accuracy of PAM in grouping music genre equal to 72.66%.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
