Rough　set　theory　is　a　mathematical　theoretic　method　to　deal　with　vague　uncertain　knowledge　and　data＇s　representation,　learning　and　induction.　But　in　the　process　of　knowledge　acquisition,　data　of　knowledge　representation　system　are　desired　to　be　precise.　The　present　arithmetic　of　knowledge　acquisition　is　inapplicable　to　vague　data.　In　this　paper　the　concept　of　rough　set　is　extended　by　idea　of　vague　set.　A　new　concept　-　vague　rough　set　is　presented.　There　are　some　discussions　about　its　operation　character　and　new　knowledge　acquisition　algorithm.　Also　in　this　paper,　an　example　of　acquiring　rules　from　vague　data　is　discussed.