Because　the　traditional　method　is　difficult　to　obtain　the　internal　relationship　and　association　rules　of　data　when　dealing　with　massive　data,　a　fuzzy　clustering　method　is　proposed　to　analyze　massive　data.　Firstly,　the　sample　matrix　was　normalized　through　the　normalization　of　sample　data.　Secondly,　a　fuzzy　equivalence　matrix　was　constructed　by　using　fuzzy　clustering　method　based　on　the　normalization　matrix,　and　then　the　fuzzy　equivalence　matrix　was　applied　as　the　basis　for　dynamic　clustering.　Finally,　a　series　of　classifications　were　carried　out　on　the　mass　data　at　the　cut-set　level　successively　and　a　dynamic　cluster　diagram　was　generated.　The　experimental　results　show　that　using　data　fuzzy　clustering　method　can　effectively　identify　association　rules　of　data　sets　by　multiple　iterations　of　massive　data,　and　the　clustering　process　has　short　running　time　and　good　robustness.　Therefore,　it　can　be　widely　applied　to　the　identification　and　classification　of　association　rules　of　massive　data　such　as　sound,　image　and　natural　resources.　©　2021,　Springer　Nature　Singapore　Pte　Ltd.