Regression based prediction of piperine content in black pepper using near infrared spectroscopy

Black pepper ( Piper nigrum L .) is a widely used spice known for its complex chemical composition and beneficial properties, including anti-inflammatory, antioxidant, anti-cholesterol, and antimicrobial effects. The spice’s distinctive aroma and flavor are largely due to its essential oil content, which is rich in bioactive compounds such as piperine, terpenes, alkaloids, flavonoids, and volatile oils. Monoterpenes like beta-pinene, sabinene, and beta-caryophyllene are significant contributors to these properties. This research aims to predict the piperine content in black pepper using near infrared (NIR) spectroscopy. The spectral data was analyzed with various regression techniques, enhanced by machine learning to assess the accuracy of predicting piperine concentration in different brand of black pepper samples from verified suppliers. NIR absorption spectroscopy, covering the wavelength range of (900–1700) nm, was employed to assess the samples. The spectral data were processed and analyzed using principal component analysis (PCA), decision trees (DT), support vector regression (SVR), extreme gradient boosting (XGBoost), random forest (RF), and partial least squares (PLS) regression. PCA was used to visualize the separation of the samples, while the other methods evaluated the fit of predictive models using metrics such as R 2 (coefficient of determination), root mean squared error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). The performance efficacy of the regression models have been evaluated using cross validation (CV), intraclass correlation coefficient (ICC), F score, and p-value. NIR spectroscopy combined with chemometric analysis effectively predicts the piperine concentration in black pepper samples and demonstrated successfully. The predicted data using NIR spectroscopy was correlated with the reverse-phase high performance liquid chromatography (RP-HPLC) measured data as reference values of black pepper samples. This approach also holds potential for application in the quality assessment of other spices and related products.