Aeroelastic analyses of bladed disks are routinely performed using Finite Element Methods (FEM) to evaluate the dynamic behavior of aircraft engine components. These analyses produce complex-valued eigenvectors, which can be transformed into real-domain mode shapes for engineering interpretation. A major post-processing challenge is the classification of the large number of modes represented in Frequency vs Nodal Diameter (FreND) diagrams, a task that is currently performed manually by domain experts and is therefore time-consuming and difficult to scale. This thesis investigates the use of Deep Learning for automated modal classification. In particular, a Convolutional Neural Network (CNN) is designed to identify modal shape labels from processed aeroelastic data and to emulate the expert labeling procedure.