“…The outcomes of this evaluation are shown in Figure 4.4. Analyzing Figure 4.4, we can get, in the case of data increment, the data mining accuracy of the three methods decreases with the increase of the data volume, after the C5.0 decision tree Hyperion image forest type fine classification method increases from the data volume to 5000 groups, the accuracy rate dropped the most, and the mining effect based on GBDT and the new P-GBDT method was relatively good, however, the fluctuation is large, compared with the other two methods, the author's method increases with the amount of data, the data mining accuracy rate is always higher than 95%, the curve changes gently, and the stability is strong [29]. Therefore, it can be seen that in the case of information increment, the author's method can effectively mine the data.…”