Imidazole-based materials have attracted considerable attention due to their promising potential for facilitating anhydrous proton transport at high temperatures. Herein, a machine learning-based deep potential (DP) model for bulk imidazole with first-principles accuracy is developed. The trained model exhibits remarkable accuracy in predicting energies and forces, with minor errors of 4.71 × 10-4 eV/atom and 3.23 × 10-2 eV/Å, respectively. Utilizing DP molecular dynamics simulations, we have systematically investigated the temperature-dependent formation and dynamics of imidazole supramolecular chains through the partial radial distribution function, quantification of hydrogen bond numbers, incoherent intermediate scattering function, and diffusion coefficient. The findings reveal the influence of temperature on the proton transport path following either the "Grotthuss" and "vehicle" mechanism.