Application of ensemble neural-network method to integrated sugar content prediction model for citrus fruit using Vis/NIR spectroscopy.

As consumers’ preference for sweetness in food products is increasing, most agricultural product processing complexes currently operate sugar-screening machines. However, every season, the entire prediction models have to be modified fit fruit species cultivated that season using a long-drawn process, because separate are required each species. Therefore, this study, species-integrated were examined based on three citrus The species-specific and performance of classical partial least squares regression (PLSR)-based, neural-network-based, ensemble-based evaluated. Four different types ensemble proposed depending combination method layers classification features. analytical results indicated Ensemble Type-4 model exhibited best both data, with average 9.9% 22.1% reduction RMSETest compared conventional PLSR methods. Furthermore, structural advantages modularity models, effective maintenance expected possible future applications field.