It　is　important　to　keep　balance　between　the　usefulness　and　the　degree　of　privacy　protection　when　the　techniques　of　generalization　and　suppression　are　applied　to　the　K-anonymization　of　data.　A　novel　K-anonymity　method　implemented　by　grey　relational　analysis　based　on　difference　information　theory　is　proposed.　The　similarity　between　any　two　data　sequences　is　described　by　their　balanced　closeness　degrees.　The　generalization　and　suppression　operations　are　carried　out　accordingly.　The　K-anonymized　data　are　then　applied　in　cluster　analysis.　The　experimental　results　on　the　real　data　sets　prove　the　effectiveness　of　the　proposed　method.