Skip to main content

Table 2 Performance comparison of our proposed model and traditional methods on dataset. The A, Se, and Sp mean Accuracy, Sensitivity, and Specificity in the table

From: Effective attention-based network for syndrome differentiation of AIDS

Methods-(%)
Syndromes Naive Bayes    SVM    RF    MLP    ATT-MLP   
  A Se Sp A Se Sp A Se Sp A Se Sp A Se Sp
S1 73.8 60.2 80.6 70.4 58.7 76.3 81.1 72.7 85.2 77.9 65.3 84.2 85.2 77.1 89.2
S2 78.9 68.3 84.3 71.7 62.3 76.4 84.5 76.5 88.5 77.6 67.5 82.7 86.4 72.2 93.4
S3 69.9 61.1 74.3 69.0 59.4 73.9 83.6 73.9 88.5 76.2 65.9 81.3 83.5 66.5 92.0
S4 79.8 72.9 83.2 72.9 63.7 77.5 85.3 81.8 87.1 82.6 74.3 86.7 87.2 79.6 91.0
S5 66.7 57.1 71.4 71.5 61.7 76.4 81.4 80.5 81.8 66.7 55.2 72.4 85.9 74.8 91.4
S6 77.2 68.4 81.6 74.5 65.6 79.0 85.2 76.7 89.5 72.8 62.6 77.9 84.2 68.0 92.2
S7 77.7 70.1 81.4 76.0 66.2 80.9 80.2 71.0 84.8 78.1 69.2 82.5 87.6 83.7 89.6
Avg. 74.8 65.4 79.5 72.3 62.5 77.2 83.0 76.1 86.5 76.0 65.7 81.1 85.7 74.6 91.3