@shinseung,
Those two labels are distinct:
- The first indicates that the sequence could not be classified beyond kingdom level (which probably indicates that it is really non-bacterial, e.g., host DNA).
- The second consists entirely of labels from the reference database you used. So QIIME is able to classify that sequence to the genus (or species?) level, but unfortunately the top match is to reference sequence(s) that are totally unannotated in the database.
I hope that clarifies!