The　increasing　use　ol　social　networks,　such　as　Facebook,　Twitter,　and　Weibo,　has　produced　and　is　producing　huge　volume　of　data.　Business　firms　and　other　organizations　are　interested　in　discovering　new　business　insight　to　increase　business　performance.　By　using　advanced　analytics,　enterprises　can　analyze　big　data　to　learn　about　relationships　underlying　social　networks　that　characterize　the　social　behavior　of　individuals　and　groups.　Using　data　describing　the　relationships,　we　are　able　to　identify　social　leaders　who　influence　the　behavior　of　others　in　the　network,　and　on　the　other　hand,　to　determine　which　people　arc　most　affected　by　other　network　participants.　This　study　focuses　on　modeling　the　knowledge　diffusion　In　social　networks.　We　will　present　a　new　evolving　model　of　a　directed,　scale-free　network.　We　will　test　the　effectiveness　of　our　model　by　a　simulation　using　data　of　a　real-world　social　network.