The machine learning based model has successfully identified 75 compounds with an accuracy range of 70-100 per cent as active compounds against SARS-CoV-2 spike protein. All the screened active compound will further investigated using molecular docking protocols and MD simulation in the next part of the work.