Predicting the Visibility of the First Crescent

Abstract

This study presents an application of machine learning to predict whether the first crescent of the lunar month will be visible to naked eye on a given date. The study presents a dataset of successful and unsuccessful attempts to find the first crescent at the start of the lunar month. Previously, this problem was solved by analytically deriving the equations for visibility parameter(s) and manually fixing threshold values. However, we applied supervised machine learning on the independent variables of the problem, and the system learnt about the criteria of classification. The system gives precision of 0.88 and recall of 0.87 and hence it treats both false positives and false negatives equally well.

Predicting new crescent moon visibility applying machine learning algorithms

“…Referring to the results obtained in 44 and 45 , which were the only studies that used some machine learning algorithms in the prediction of the new crescent Moon, the comparison of the results is irrational. The dataset in 45 was restricted to only one country and employed only one ML algorithm.…”

“…The dataset in 45 was restricted to only one country and employed only one ML algorithm. However, the study in 44 used dataset different from ours; our dataset is more accurate and complete, as mentioned earlier. Both datasets are imbalanced.…”

“…Tafseer 44 considered the problem of Moon crescent visibility for each month as a classification problem and suggested the use of ML algorithms instead of mathematical or astronomical methods. His study was based on the dataset found on the website managed by ICOP.…”

“…In an effort to help in this matter, this current study suggests the application of ML algorithms in the prediction of the visibility of the new crescent Moon. To the best of our knowledge, in the context of this research, only 44 and 45 tackled the prediction of the crescent Moon visibility applying some ML algorithms. Nevertheless, Allawi 45 was confined to a single ML algorithm, and it was just restricted to one individual country, Iraq.…”

“…Nevertheless, Allawi 45 was confined to a single ML algorithm, and it was just restricted to one individual country, Iraq. Referring to the dataset used in 44 , several limitations were identified such as DAZ feature which should be positive numbers, the values of some features like ARCL (elongation) were not accurate, names of some countries and cities features were not consistent; same countries and cities in different languages or abbreviated, which may affect the analysis results. To overcome all these shortcomings, we built our own dataset and included all the years up to the current year 2022, as explained in " The proposed model " section.…”