Reduct　finding,　especially　optimal　reduct　finding,　similar　to　feature　selection　problem,　is　a　crucial　task　in　rough　set　applications　to　data　mining.　In　this　paper,　we　have　studied　the　basic　concepts　of　rough　set　theory,　and　discussed　several　special　cases　of　the　Ant　Colony　Optimization　metaheuristic　algorithms.　Based　on　the　above　study,　we　propose　a　feature　selection　algorithm　within　a　mixed　framework　based　on　rough　set　theory　and　Ant　Colony　Optimization.　Experimental　results　show　that　the　algorithm　of　this　paper　is　flexible　for　feature　selection.