In this work, we utilized a data-driven approach to examine the nature of the doubly charmed \(T_{cc}(3875)^{+}\) tetraquark. A large ensemble of synthetic scattering amplitudes were generated based on the uniformized \(\mathcal {S}\) -matrix framework. We trained 50 deep neural network classifiers to recognize subtle differences in the line shapes that arise from distinct pole topologies. When the LHCb data was fed to the classifiers, the trained DNN models consistently identify an isolated pole in the [bt] sheet which is indicative of a hadronic molecular nature for the \(T_{cc}(3875)^{+}\) .
