Skip to main content

Table 13 Performance of BioHEL on learning CN and RSA using the reduced PSSM representation

From: Automated Alphabet Reduction for Protein Datasets

Dataset Representation Accuracy #Rules #expr. atts. Run-time(h)
CN PSSM 81.4 ± 0.4 269.5 ± 12.7 14.4 ± 3.1 31.4 ± 2.9
  Reduced PSSM 80.8 ± 0.3 256.0 ± 12.1 12.9 ± 2.7 19.4 ± 1.7
SA PSSM 78.5 ± 0.4 520.4 ± 15.3 13.9 ± 3.5 54.3 ± 4.2
  Reduced PSSM 77.6 ± 0.4 493.5 ± 14.9 12.9 ± 3.3 34.3 ± 3.0