For polymer composites, nanocomposites and polymer thin film systems, the local properties of polymers can be altered by the chemical and physical interactions with substrates and embedded particles over a length scales exceeding 100nm. The mechanisms and impact of confined polymers remains still an active area of research and debate. Here we will review methods to explore nanoscale polymer properties near surfaces, with a focus on scanning probe methods to quantitatively measure mechanical response and the interesting mechanics problems that arise. In multiphase soft materials, local changes in the sample modulus, tip-sample interactions and stress field interaction effects impact the acquired force curves. Coupling experimental data with simulations of indentations enable the structural effects of the particle-polymer-tip system to be accurately estimated and removed, revealing the effects of confinement on property gradients. Capturing and archiving this data allows case studies which connect the property-structure-property domains through a combination of machine learning and physics-based modeling. We demonstrate the ability to identify the most critical features influence properties and the ability to acquire new insights from ensembles of unrelated data. The importance of data, data resources and leverage of this knowledge in new physics based and interpretable machine learning methods is discussed. Overall this work illustrates new approaches combining physics and data based models and experiments to tackle materials design principles for the complex, high dimensional problems inherent in the multi-phase polymer space.