As Sarofi, Muhammad Abid (2020) Klasifikasi Genre Musik Berdasarkan Mel Frequency Cepstrum Coefficient (MFCC) dengan Menggunakan Metode Support Vector Machine (SVM) dan Random Foret (RF). Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211640000082-Undergraduate_Thesis.pdf - Accepted Version Download (2MB) | Preview |
Abstract
Genre musik adalah pengelompokan musik sesuai dengan kemiripan antara satu musik dengan musik yang lainnya, seperti kemiripan dalam hal frekuensi musik, struktur ritmik, dan konten harmoni. Genre musik merupakan hal yang penting dalam pengelompokkan musik. Pengelompokkan tersebut dilakukan secara manual pada umumnya dengan cara mendengarkan secara langsung lagu tersebut. Namun, hal tersebut dapat menimbulkan ketidakefisiensian dalam mengelompokkan lagu. Tujuan dari penelitian ini adalah untuk mendapatkan metode yang memiliki performa klasifikasi terbaik diantara dua metode yang digunakan yaitu SVM dan Random Forest, dimana metode yang terbaik nantinya akan digunakan dalam pembuatan GUI. Fitur ekstraksi yang digunakan dalam penelitian ini adalah MFCC, karena MFCC mampu mengadaptasi pendengaran manusia. Hasil dari klasifikasi audio pada GTZAN dataset dengan menggunakan kedua metode menunjukkan bahwa Random Forest merupakan metode yang terbaik daripada Support Vector Machine karena memiliki nilai accuracy, precision, sensitivity dan Fscore yang lebih besar.
===================================================================================================================================
A music genre is a category that identifies some pieces of music as belonging to a share tradition or set of conventions. Music can be divided into different genres in many different ways, such as similarity in terms of music frequency, rhytmic structure, and harmony content. Nowdays, companies use music classification to be able to give recommendations to their customers. The first step in that direction is determining music genres. Usually, the music genre categorizing is done manually by listening directly to the music. However, inefficiency became a problem that must be considered in doing music classification manually, because it will take a lot of time. Therefore, it is needed to conduct a study based on machine learning to classify music genres. In this study, two methods i.e. Support Vector Machine (SVM) and Random Forest are compared to classify music clips into different genres, whereas the best method are use to make the GUI. Mel Frequency Cepstrum Coefficient (MFCC) is used for feature extraction because it is easy to implement, robust to noise and represent frequencies that can be captured by the human ear. The study conducted on GTZAN dataset shows that audio classification using Random Forest has a higher accuracy, precision, sensitivity and Fscore than SVM.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSSt 519.53 Ass k-1 2020 |
| Uncontrolled Keywords: | MFCC, Genre Musik, Random Forest, SVM |
| Subjects: | H Social Sciences > HA Statistics Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) Q Science > QA Mathematics > QA76.9.U83 Graphical user interfaces. User interfaces (Computer systems)--Design. |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Muhammad Abid As Sarofi |
| Date Deposited: | 13 Mar 2025 06:31 |
| Last Modified: | 13 Mar 2025 06:31 |
| URI: | http://repository.its.ac.id/id/eprint/73655 |
Actions (login required)
 |
View Item |