Research Article

Application of Stochastic Gradient Boosting Approach to Early Prediction of Safety Accidents at Construction Site

Table 3

Comparison of prediction results of SGB, ANN, and DT models.

No.TargetSGB model predictionANN model predictionDT model prediction
1st2nd3rd1st2nd3rd1st2nd3rd

15563524251
26563524162
32231124231
41316164123
51513154132
63362412231
73213142231
84423142231
95451524162
104342142231
112431412431
122231214231
135561154152
141412421231
151213124123
163465412231
174214142123
183213123231
192156524213
205154214162
211513241213
222564254213
231123214231
241625164123
256213241231
264432124143
271123125123
284416124231
294412146143
301431162143
Accumulated accuracy1st1st-2nd1st-2nd-3rd1st1st-2nd1st-2nd-3rd1st1st-2nd1st-2nd-3rd
36.6% (11/30)63.3% (19/30)76.6% (23/30)33.3% (10/30)60.0% (18/30)80.0% (24/30)33.3% (10/30)63.3% (19/30)70.0% (21/30)