From: Predicting anatomic therapeutic chemical classification codes using tiered learning
2 nd level ATC Code Prediction using different initializations and comparison to STP | ||||||
---|---|---|---|---|---|---|
Initialization method | Supervised learning | ChEMBL (Perfect 1st level) | SuperPred Prediction Results | |||
Prediction method | A(2) | CA(2) (N(2)) | A(2) | CA(2) (N(2)) | A(2) | CA(2) (N(2)) |
MLP | 50.00% | 52.21% (43.63%) | 65.20% | 82.60% (53.39%) | 44.61% | 43.38% (39.22%) |
Naive Bayes | 41.67% | 37.35% (25.00%) | 51.47% | 75.74% (51.47%) | ||
Random Forest | 49.51% | 50.98% (42.16%) | 64.71% | 82.35% (65.20%) | ||
SVM | 52.45% | 53.92% (45.10%) | 67.16% | 83.58% (67.65%) |