Greenhouse　gas　emissions　are　widely　recognized　as　the　primary　cause　of　global　warming,　leading　to　a　growing　attention　on　carbon　emission　management.　However,　the　existing　studies　still　failed　to　propose　a　feasible　approach　to　directly　forecast　carbon　emission　trends　and　also　did　not　take　into　account　both　environmental　regulation　and　efficiency　improvement.　Hence,　this　study　aims　to　propose　a　novel　carbon　emission　trend　forecast　model　based　on　data-driven　rule-base　with　considering　the　intensity　coefficient　of　environmental　regulation　and　the　management　efficiency　of　carbon　emissions.　Carbon　emission　data　of　30　Chinese　provinces　are　collected　to　illustrate　the　effectiveness　of　the　proposed　model.　Results　indicated　that:　1)　the　data-driven　rule-base　model　is　able　to　directly　forecast　carbon　emission　trends　within　range　from　-18.54　%　to　19.18　%;　2)　by　integrating　regulation　intensity,　the　predicted　results　of　the　model　have　smaller　carbon　emission　tends,　e.g.,　decrease　of　average　changing　rate　from　0.4100　to　0.2762;　3)　by　further　integrating　efficiency　improvement,　the　predicted　results　align　more　with　the　expected　objectives　of　policy　makers,　i.e.,　the　average　carbon　emission　efficiency　approximates　0.8920　and　the　number　of　provinces　being　effective　efficiency　is　increased　to　8.　These　findings　also　highlighted　the　importance　of　carbon　emission　tend　forecast　with　environmental　regulation　and　efficiency　improvement.　The　proposed　carbon　emission　trend　forecast　model　could　serve　as　an　alternative　tool　for　achieving　dual　carbon　goals　in　the　context　of　China.