Classificatⅰ on results Average accuracy Model Discrete Embedding Settings BOW BOW+ENTITY Word2vec 20NG-SIM 90.81% 91.11% 91.67% Nikolov 2013 20NG-DIE 96.66 9690% 98.27% Window 5 GCAG-SIM 94.15% 94.29 9681% Dim: 400 GCAT-DIF 8898% 90.18% 90.64% Average accuracy Model SVMHIN SVMHIN+KnowSim IndefSVMHIN+KnwoSim Settings DWD DWD+other Dwd DWD+other Meta paths Metapaths 20NGSM91.60% 92.32% 92.68 92.65% 93.38% 20NGDF97.20% 97.83% 98.01 98.13 9845% GCAG-SIM.82% 95.29 96.04% 95.63 98.10% GCAT-DIF.19% 90.70 91.88 91.63% 93.51% Collective classification: Lu and gator 2003 Kong et al. 2012 52Average accuracy Model SVMHIN SVMHIN+KnowSim IndefSVMHIN+KnwoSim Settings DWD DWD+other MetaPaths DWD DWD+other MetaPaths 20NG-SIM 91.60% 92.32% 92.68% 92.65% 93.38% 20NG-DIF 97.20% 97.83% 98.01% 98.13% 98.45% GCAG-SIM 94.82% 95.29% 96.04% 95.63% 98.10% GCAT-DIF 91.19% 90.70% 91.88% 91.63% 93.51% Classification Results Average accuracy Model Discrete Embedding Settings BOW BOW+ENTITY Word2vec 20NG-SIM 90.81% 91.11% 91.67% 20NG-DIF 96.66% 96.90% 98.27% GCAG-SIM 94.15% 94.29 96.81% GCAT-DIF 88.98% 90.18% 90.64% Collective classification: Lu and Gatoor 2003; Kong et al. 2012 Mikolov 2013. Window: 5 Dim: 400 52