Sentinel 2 is a great tool for land cover classification.
If you want to get an idea of how to start doing it, I would reccomend the SCP plugin for qgis, through which you can easely access the data and preprocess them, before moving to the classification. See https://semiautomaticclassificationmanual-v5.readthedocs.io/it/latest/tutorial_2.html
a combination of Sentinel-1 and Sentinel-2 images increase the number of layers for the classification algorithm. Sentinel1 (which has polarization VV and VH) is very sensitive to free water surface and sharp edge features (buildings). I guess you can use a combination of S1 and S2 in SCP, as suggested by Federico Amato