Coffee classification involves the differentiation of coffee based on species, geographical origin, and quality attributes. In this study, ground coffee samples from seventeen commercial brands: Mai, Honey, Duong, M Ja, Con Soc, Hat A, Trung Nguyen, Huong Viet, Vina, Phuc Long, TwitterBean, Nhan, Phuong Vi, Thu Ha, Nes, and Highland were analyzed using Attenuated Total Reflectance Fourier Transform Infrared (ATR-FTIR) spectroscopy. Spectra were acquired in transmittance mode over the range 400-4000 cm-1 with 128 scans per sample and a resolution of 10 cm-1. Sample surfaces were smoothed consistently during measurement to minimize variability. The resulting spectral data were compressed and subjected to multivariate statistical analysis, including Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). These techniques were applied to classify four roasted coffee types: Arabica, Robusta, Moka, and Culi. PCA revealed that the first two principal components accounted for 72.99% and 11.97% of the total variance, respectively. While PCA enabled partial brand-level discrimination, LDA provided clear separation among all four coffee types. Our findings demonstrate that ATR-FTIR spectroscopy, when coupled with chemometric modeling, offers a rapid and robust alternative to conventional methods for coffee authentication. This approach enables effective classification of Vietnamese coffee products and holds promise for future applications in fraud detection and quality control.