A new comparative study based on applying two-way and three-way data analysis methods to the fluorescence excitation-emission matrices (EEMs) was performed to co-estimate two active pharmaceutical ingredients (amlodipine besylate, AD, and telmisartan, TS) in a binary mixture. Two-way data analysis was carried out by a bilinear model algorithm using multivariate curve resolution-alternative least squares (MCR-ALS). Three-way data analysis was performed by trilinear models using parallel factor analysis (PARAFAC), and Tucker3. Loadings or profiles obtained by applying the MCR-ALS, PARAFAC, and Tucker3 to the EEMs in the presence of the overlapping excitation and emission spectra made it possible to obtain the individual contribution of the target drugs without requiring a chromatographic separation methodology. The excitation-emission spectra of the analytes were identified from the excitation and emission profiles. The quantity of two pharmaceuticals in analyzed samples was obtained from the relative concentration profiles using univariate calibration. The calibration curves were found to be linear in the range of 0.5–5.0 μg/mL for both drugs. The validity of the methods was verified by analyzing validation samples. Then, the applicability of MCR-ALS, PARAFAC, and Tucker3 approaches was tested for the assay of AD and TS in commercial tablets. For a comparison of the proposed bilinear and trilinear approaches, the statistical analyses including the one-way ANOVA and Bartlett tests were applied to the means and variances of the assay results, respectively. The statistical results showed that the precision and accuracy of the experimental results obtained with the MCR-ALS, PARAFAC, and Tucker3 models were comparable for the assay of the two target drugs.