Application of nirs technology and machine learning for predicting chlorogenic acid in arabica coffee beans <i>(Coffea arabica L.)</i>

Abstract The Near-Infrared Spectroscopy (NIRS) technology, combined with machine learning, offers a promising approach for non-destructive analysis of agricultural products. This study focuses on predicting chlorogenic acid content in Arabica coffee beans using NIRS and machine learning models. Various preprocessing methods were applied to enhance the accuracy of the model, including peak normalization, Savitzky-Golay smoothing, and extended multiplicative scatter correction (EMSC). The results show that the model can effectively predict chlorogenic acid levels with high accuracy, especially when using Savitzky-Golay smoothing as a preprocessing method. The model achieved a coefficient of determination (R 2 ) of 0.79 and a ratio of prediction to deviation (RPD) of 3.38, indicating a robust and reliable prediction. These findings underscore the potential of integrating NIRS and machine learning for quality control in the coffee industry.