Determining the Capability of the Tree-Based Pipeline Optimization Tool (TPOT) in Mapping Parthenium Weed Using Multi-Date Sentinel-2 Image Data

The Tree-based Pipeline Optimization Tool (TPOT) is a state-of-the-art automated machine learning (AutoML) approach that automatically generates and optimizes tree-based pipelines using genetic algorithm. Although it has been proven to outperform commonly used techniques, its capability handle high-dimensional datasets not investigated. In vegetation mapping analysis, multi-date images are generally contain embedded information, such as phenological canopy structural properties, known enhance accuracy. However, without the implementation of robust classification algorithm or feature selection tool, large sets presence redundant variables in can impede accurate efficient landscape classification. Hence, this study sought test efficacy TPOT on Sentinel-2 image optimize accuracies infested by noxious invasive plant species, parthenium weed (Parthenium hysterophorus). Specifically, models created from image, an system combines TPOT, dubbed “ReliefF-Svmb-EXT-TPOT”, were compared. results showed could perform well data with sets, but at computational cost. overall 91.9% 92.6% ReliefF-Svmb-EXT-TPOT models, respectively. findings crucial for geospatial limited human intervention.