Biosecurity and biosafety are key concerns of modern society. Although nanomaterials are improving the capacities of point detectors, standoff detection still appears to be an open issue. Laser-induced fluorescence of biological agents (BAs) has proved to be one of the most promising optical techniques to achieve early standoff detection, but its strengths and weaknesses are still to be fully investigated. In particular, different BAs tend to have similar fluorescence spectra due to the ubiquity of biological endogenous fluorophores producing a signal in the UV range, making data analysis extremely challenging. The Universal Multi Event Locator (UMEL), a general method based on support vector regression, is commonly used to identify characteristic structures in arrays of data. In the first part of this work, we investigate fluorescence emission spectra of different simulants of BAs and apply UMEL for their automatic classification. In the second part of this work, we elaborate a strategy for the application of UMEL to the discrimination of different BAs’ simulants spectra. Through this strategy, it has been possible to discriminate between these BAs’ simulants despite the high similarity of their fluorescence spectra. These preliminary results support the use of SVR methods to classify BAs’ spectral signatures.