Usuwanie duplikatów wierszy z Oracle

delete from employee a
where employee_id in (
   select employee_id 
   from employee b
   where b.department_id > a.department_id )