Utilization of Machine Learning Approaches for Rainfall Data Imputation: A Systematic Literature Review

Incomplete data is a frequent issue in rainfall observation data records which can affect the results of analysis or modeling using rainfall data. Incompleteness of rainfall data can occur due to various obstacles that may arise in measuring instrument operations. Hence, to handle incomplete rainfall data in a precise manner, numerous machine learning techniques have been utilized to impute missing rainfall data with reliable values. This article offers a structured analysis, not only various methods employed for filling in the gaps of missing rainfall data but also the data characteristics, and measurement factors. The main purpose of this review paper is to highlight the possible enhancements that can be made to current machine learning models and help readers gain a better understanding of the prevailing trends in imputation techniques for estimate the missing rainfall data.