Table 4 Comparative evaluation of Theme and LDA-Naïve clusters on the SNP dataset using precision (P), recall (R), and F-score (F) metrics
From: Discovering themes in biomedical literature using a projection-based algorithm
# Cl | Method | P | R | F |
|---|---|---|---|---|
100 | LDA-Naïve-100 | 0.358 | 0.364 | 0.361 |
100 | Theme uni | 0.688 | 0.302 | 0.419 |
587 | LDA-Naïve-1000 | 0.507 | 0.278 | 0.359 |
587 | Theme uni | 0.639 | 0.226 | 0.334 |