Learning Desirable Matchings From Partial Preferences