Skip to main content

Advertisement

Table 4 Evaluation of learning model

From: Database application model and its service for drug discovery in Model-driven architecture

Dataset No. Clustering Correct rate (%) Dataset No. Clustering Correct rate (%)
1 41 –100 % 100 5 34.7 –100 % 100
  0.1 –100 % 100   27.4 –100 % 99.5
2 55 –100 % 23   12.1 –100 % 99.6
  29 –100 % 95   0.1 –100 % 100
  9 –100 % 40 6 53.2 –100 % 100
  0.1 –100 % 100   51 –100 % 100
3 70.3 –100 % 100   50 –100 % 100
  36 –100 % 90   34.7 –100 % 100
  18 –100 % 20   27.4 –100 % 98
  0.1 –100 % 86   18 –100 % 98
4 81.7 –100 % 100   12.1 –100 % 99
  63.5 –100 % 100   0.1 –100 % 100
  34.7 –100 % 100 7 81.7 –100 % 100
  27.4 –100 % 100   49 –100 % 100
  12.1 –100 % 100   18 –100 % 99
  0.1 –100 % 100   0.1 –100 % 100
5 94 –100 % 100    
  1. “Dataset No”. indicates 7 dataset patterns of pinpoint data shown in Table 2. “Clustering” indicates the approximated interval of prediction in side effect incidence of drug. “Correct rate” indicates to what extent each example of learning model can meet each correct interval in the prediction