Analytic predictive of crescent sighting using astronomical data-based multinomial logistic regression in Indonesia

Sugiharto, Tomy Ivan, Hariyadi, M. Amin ORCID: https://orcid.org/0000-0001-9327-7604, Chamidy, Totok, Santoso, Irwan Budi and Crysdian, Cahyo ORCID: https://orcid.org/0000-0002-7488-6217 (2025) Analytic predictive of crescent sighting using astronomical data-based multinomial logistic regression in Indonesia. G-Tech Jurnal Teknologi Terapan, 9 (4). pp. 2240-2247. ISSN 2623-064X

Preview

Text 28032.pdf - Published Version Available under License Creative Commons Attribution. Download (382kB) | Preview

Abstract

This research aims to develop and validate a sophisticated crescent visibility classification model in Indonesia. Multinomial Logistic Regression (MLR) was chosen for its capability to provide clear model interpretation through coefficient analysis. Utilizing comprehensive observational data (2021-2025) from Indonesia's Meteorology, Climatology, and Geophysics Agency (BMKG), the study comprised 2210 data points. The model classifies visibility into three categories (Dark, Faint, and Bright) based on defined elongation thresholds. The final predictor variables used were azimuth difference, moon altitude, and elongation. Analysis of the optimal model's (Model A3) coefficients revealed azimuth difference and elongation as the most dominant predictors, marked by exceptionally large positive coefficients (12.050 and 12.018, respectively) for classifying the 'Faint' category. After data preprocessing and systematic optimization ('saga' solver, L2 penalty), the optimal model (A3, C=100) demonstrated exceptional performance with an outstanding F1-Score of 99.10%. These findings strongly validate MLR's effectiveness for elongation-based crescent visibility classification and highlight its substantial potential as a reliable foundation for objective decision-making.

Item Type:	Journal Article
Keywords:	classification; elongation; hilal visibility;multinomial logistic regression.
Subjects:	08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080105 Expert Systems 08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080199 Artificial Intelligence and Image Processing not elsewhere classified
Divisions:	Graduate Schools > Magister Programme > Graduate School of Informatics Engineering
Depositing User:	Mokhamad Amin Hariyadi
Date Deposited:	13 Jul 2026 11:12

Downloads

Origin of downloads

Actions (login required)

View Item