This Author published in this journals
All Journal Technomedia Journal
I Made Ade Prayoga
Universitas Pendidikan Ganesha

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Pengelompokan Laras Suara Berdasarkan Pepatutan Atau Pathet Gamelan Bali Menggunakan Klasifikasi K-Nearest Neighbor Dan Support Vector Machine I Made Ade Prayoga; Gede Indrawan; Dewa Gede Hendra Divayana
Technomedia Journal Vol 8 No 2 Special Issues (2023): Special Issue: Sistem Informasi Manajemen Dalam Menunjang Teknolog
Publisher : Pandawan Incorporation, Alphabet Incubator Universitas Raharja

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33050/tmj.v8i2SP.2011

Abstract

Gamelan is an orchestra consisting of instruments made of stone, wood, bamboo, iron, bronze, leather, strings and others using pelog and slendro barrels, and has 7 pepatutan or pathet namely; (1) pathet selisir, (2) pathet panji, (3) pathet tembung (4) pathet sunaren, (5) pathet baro, (6) pathet pengenter, dan (7) pathet malat, each pepatutan or pathet has special characteristics with the rules for how to play it in each Balinese Gamelan group. Along with the development of the era, there is a transition in the way of teaching in the past and now that is different, so that today's children only know the order in which the gamelan blades are struck, not the barrel of the gamelan sound. Therefore, the author wants to build a system that can classify sound tunings into 7 pepatuans or pathets contained in the Balinese Gamelan. This system will be designed and built based on the appropriate grouping or pathet obtained using the K-Nearest Neighbor algorithm and Support Vector Machine. Based on the test results, the KNN algorithm gives more effective results in grouping sound barrels with a percentage accuracy rate of 100%, while the SVM algorithm gives an accuracy percentage of 74.29%. Testing of the time required in the classification process also shows that KNN provides a faster processing time of 0.14388 seconds compared to SVM, which is 0.17642 seconds. KNN gives better results because, in principle, K-NN chooses the nearest neighbor which uses the distance parameter, namely the Euclidean distance, which is very suitable for use in determining the shortest distance between two data.