An Improved Methodology for Filling Missing Values in Spatiotemporal Climate Dataset: Application to Tanganyika Lake Dataset

摘要

In this paper, an improved methodology for the determination ofmissing values in a spatio-temporal database is presented. This methodologyperforms denoising projection in order to accurately fill the missing valuesin the database. The improved methodology is called EOF Pruning and itis based on an original linear projection method called Empirical Orthogo-nal Functions (EOF). The experiments demonstrate the performance of theimproved methodology and present a comparison with the original EOF andwith a widely-used Optimal Interpolation method called Objective Analysis.