Claim Missing Document
Check
Articles

Found 1 Documents
Search

Implementasi Metode Modified K-Nearest Neighbor untuk Klasifikasi Status Gunung Berapi Fikar Cevi Anggian; Nurul Hidayat; Muhammad Tanzil Furqon
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 12 (2019): Desember 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (301.289 KB)

Abstract

Indonesia has numerous disaster-prone area, therefore, it is known as a 1001 disaster country. Natural disasters are inevitable, nevertheless, the impact can be minimized with proper anticipation. One of the disasters that often occur in Indonesia is volcanic eruption. Indonesia has 127 active volcanoes that are ready to erupt anytime. Indonesia is also known to contribute to around 30% of the world's volcanoes, and they are located near the residential areas. Casualties are often found in every volcanic eruption due to lack of anticipation from residents who live nearby the volcano. To minimize life and material loss, early warning is needed to provide quick and accurate notification of the volcano. This research used the Modified K-Nearest Neighbor method to classify volcano status. The data used are 110 data obtained from the official government agency that authorized to issue volcanic status, known as Pusat Vulkanologi dan Mitigasi Bencana Geologi (PVMBG). The test was carried out using various k values, namely 3,5,6,7, and 9. The highest accuracy obtained in this research was 86.87%, and the average accuracy was 82.87%.