Table 3 Comparison of PDD performance in class association and entity clustering before and after auto-error-correction.
From: Theory and rationale of interpretable all-in-one pattern discovery and disentanglement system
Accuracy summary | Before class readjusting | After class readjusting |
|---|---|---|
(a) Classification (based on class association) | ||
Cytochrome c | 94.56% | 97.83% |
Scavenger Class A APC | 96.84% | 97.89% |
Breast Cancer | 95.42% | 99.86% |
Heart Disease | 94.81% | 100% |
(b) Entity Clustering | ||
Cytochrome c | 97.83% | 98.91% |
Scavenger Class A APC | 95.70% | 95.70% |
Breast Cancer | 97.29% | 99.28% |
Heart Disease | 80.00% | 86.67% |
