1. Dimensionality Reduction and Similarity Metrics for Predicting Crop Yields in Sparse Data Microclimates

This study explores and develops new methodologies for predicting agricultural outcomes, such as crop yields, in microclimates characterized by sparse meteorological data. Specifically, it focuses on reducing the dimensionality in time series data as a preprocessing step to generate simpler and more explainable forecast models. Dimensionality reduction helps in managing large data sets by simplifying the information into more manageable forms without significant loss of information. We explore... L. Huender, M. Everett