Komang Nopa Sudarma
Universitas Dipa Makassar

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

Found 1 Documents
Search

Earthquake Prediction in Indonesia using Descriptive Statistics, Pearson Correlation, and Ensemble Machine Learning (Random Forest, XGBoost, LightGBM) Nunu Ariatmi; Alwin Sande; Komang Nopa Sudarma; Rismayani Rismayani
Journal of Information Technology and Its Utilization Vol 9 No 1 (2026): May 2026
Publisher : Sekolah Tinggi Multi Media

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56873/jitu.9.1.6060

Abstract

Indonesia is a seismically active region because it is crossed by the meeting point of three tectonic plates: the Indo-Australian Plate, the Eurasian Plate, and the Pacific Plate, commonly referred to as the Pacific Ring of Fire. This research seeks to forecast earthquake occurrences in Indonesia by integrating descriptive statistics, Pearson correlation analysis, and ensemble-based machine learning techniques including Random Forest, XGBoost, and LightGBM. The dataset is sourced from the BMKG and covers the period from September 1 to December 20, 2024. The methods used include descriptive statistics, which can help identify trends and patterns in earthquake data such as frequency, average magnitude, and geographical distribution; Pearson correlation to show the relationship between earthquake variables such as magnitude, depth, and location; and ensemble machine learning to help predict the likelihood of earthquakes based on historical data patterns. The use of descriptive statistics, Pearson correlation, and ensemble machine learning in earthquake prediction is an important step toward enhancing understanding of earthquakes and aiding in disaster risk mitigation efforts