Abstract:
Various honey samples and possible adulterants has been characterized using Fourier Transform Infrared (FT-IR) spectroscopy integrated with ATR sampling. Spectral data of twelve varieties of samples including mono-floral honey, multi-floral honey and different variety of adulterants has been collected in Mid-IR region (4000cm -1 -400cm -1 ). Spectral Mid-IR data has been corrected using baseline correction method and preprocessed using 2nd order derivative & Standard Normal Variate (SNV) methods for removal of any additive & multiplicative scattering effects. The principal component analysis (PCA) has been used for dimension reduction in the data set and exploratory data analysis of the honey and its adulterant samples. K-means, K-medians and Fuzzy C Means based classification model has been developed for the classification of pure honey samples and adulterants. Developed model has been cross-validated using external samples.