Citation: Marina Popolizio, Alberto Amato, Federico Liquori, Tiziano Politi, Alessandro Quarto, and Vincenzo Di Lecce, "The GAIN Method for the Completion of Multidimensional Numerical Series of Meteorological Data" IAENG International Journal of Computer Science, vol. 48, no.3, pp496-506, 2021
Abstract: The missing data imputation is a very significant topic which captures considerable interest, given the importance it has in many applications. This paper analyzes the use of GAIN (Generative Adversarial Imputation Networks) to address the problem of missing data in meteorological data sets. A detailed description of the numerical method is given together with a MATLAB implementation which will be available on request. Numerical tests are presented to validate the effectiveness of this method; moreover, a comparison on a real dataset is done with the commonly used ARMA method and GAIN turns out to be more accurate.
Keyword: Artificial Intelligence, Missing Data, Imputation, Neural Network, GAIN