In　order　to　determine　the　grade　of　spontaneous　combustion　tendency　of　sulfide　ores　in　sulfur-rich　ore　deposits　scientifically,　three　determination　indexes　were　selected,　including　the　oxygen　absorption　rate　on　average,　self-heating　point,　and　ignition　point.　The　classification　model　of　spontaneous　combustion　tendency　of　sulfide　ores　was　established　based　on　entropy　weight　and　the　cloud　model.　The　determination　index　of　spontaneous　combustion　tendency　of　sulfide　ores　and　the　classification　standard　were　transformed　into　normal　cloud　classification　standard.　The　index　weight　of　actual　test　value　for　each　index　were　empowered　by　entropy　method.　The　level　of　spontaneous　combustion　tendency　of　each　sample　was　determined　based　on　the　principle　of　maximum　membership　degree　of　fuzzy　subset　B.　The　feasibility　of　this　model　was　confirmed　by　using　10　representative　samples　from　a　typical　mine　in　China.　The　classification　results　were　consistent　with　that　obtained　by　the　method　of　multi-factor　comprehensive　evaluation.　The　model　can　increase　the　accuracy　effectively　in　determining　the　level　of　spontaneous　combustion　tendency　of　sulfide　ores,　so　it　provides　a　new　method　for　controlling　the　spontaneous　fire　of　sulfur-rich　mines.　©　2018　Science　Press.　All　rights　reserved.