Accelerating the innovation process in material development requires the quick and effi-cient identification of suitable structural materials. High-throughput methods offer a promising approach in this regard, capturing easily measurable quantities, called descriptors, that are physically or statistically correlated with known material properties. This project focuses on advancing the laser-induced shockwave indentation (LiSE) testing method. In this technique, a TEA-CO₂ laser reproducibly generates shockwaves that drive an indenter into the material sample within microseconds. The aim is to significantly in-crease the throughput of this method. To achieve this, a data-driven evaluation methodology based on machine learning will be developed to enable the inference of relevant material parameters from measurement data. First, new and meaningful descriptors will be identified and validated. Indentation behaviour will be modelled as a contact problem and simulated using a robust mixed fini-te element method (FEM), with the material parameters to be determined integrated di-rectly into the simulation. Efficient, parallelised solvers will be employed to accelerate computation. Using locally refined FE meshes increases simulation accuracy. These simu-lations can be used to determine unknown material properties through numerical para-meter identification. To accelerate the computationally intensive determination of mate-rial parameters from experimental data further, neural networks will be used that have been trained on the simulation results. Furthermore, various neural network-based ap-proaches for analysing the measurement data will be developed and compared in terms of accuracy and computational efficiency.