Sentence-level words and phrases that predict mythic vs. historical content
The following metrics are from the logistic regression classifier's performance on the test set:
| Class | Precision | Recall | F1-Score | Support |
|---|---|---|---|---|
| Historical | 0.777 | 0.780 | 0.779 | 255 |
| Mythic | 0.634 | 0.630 | 0.632 | 154 |
| Overall Accuracy | 0.724 | 409 | ||
| Word/Phrase | Coefficient | Mythic Count | Non-mythic Count | p-value | q-value |
|---|---|---|---|---|---|
| θησεὺς | 2.3105 | 26 | 1 | 6.04e-11 | 6.04e-08 |
| εἶναι | 2.0600 | 82 | 49 | 4.48e-10 | 2.24e-07 |
| λέγουσιν | 1.9842 | 52 | 25 | 2.03e-08 | 4.06e-06 |
| θησέως | 1.9725 | 24 | 2 | 3.71e-09 | 1.1e-06 |
| παῖδα | 1.7512 | 36 | 13 | 1.09e-07 | 1.36e-05 |
| ἰλίου | 1.6861 | 12 | 0 | 5.48e-06 | 0.000422 |
| ὑπὸ | 1.6794 | 90 | 83 | 1.58e-05 | 0.000928 |
| φασιν | 1.6274 | 25 | 7 | 1.33e-06 | 0.000148 |
| θησέα | 1.5370 | 19 | 0 | 4.42e-09 | 1.1e-06 |
| ὄνομα | 1.5124 | 32 | 14 | 4.51e-06 | 0.000376 |
| αὐτὸν | 1.4947 | 39 | 24 | 3.58e-05 | 0.00171 |
| αὐτὴν | 1.4858 | 29 | 16 | 0.000131 | 0.00467 |
| ἀχιλλέως | 1.4736 | 16 | 0 | 9.41e-08 | 1.34e-05 |
| τὸν | 1.4653 | 165 | 190 | 2.72e-05 | 0.00143 |
| γενέσθαι | 1.3630 | 36 | 25 | 0.000315 | 0.00872 |
| εἰσιν | 1.3227 | 16 | 17 | 0.161 | 0.478 |
| ὅμηρος | 1.3213 | 15 | 1 | 2.75e-06 | 0.00025 |
| λέγουσι | 1.2954 | 31 | 14 | 8.97e-06 | 0.000641 |
| ἀποθανεῖν | 1.2744 | 23 | 3 | 5.44e-08 | 9.06e-06 |
| λέγεται | 1.2107 | 29 | 16 | 0.000131 | 0.00467 |
| ἡρακλέους | 1.1844 | 17 | 2 | 2.53e-06 | 0.00025 |
| κάθηται | 1.1660 | 6 | 0 | 0.00238 | 0.0396 |
| πρῶτον | 1.1588 | 22 | 9 | 9.13e-05 | 0.0038 |
| μνῆμά | 1.1539 | 10 | 3 | 0.00621 | 0.069 |
| ἡρῷον | 1.1354 | 9 | 1 | 0.000772 | 0.0184 |
| λόγον | 1.1086 | 12 | 3 | 0.000738 | 0.018 |
| ἀλκάθου | 1.0890 | 7 | 0 | 0.000866 | 0.0197 |
| φασὶν | 1.0742 | 14 | 7 | 0.00503 | 0.0621 |
| μάχην | 1.0490 | 10 | 6 | 0.0353 | 0.207 |
| τὸ ὄνομα | 1.0425 | 17 | 4 | 3.86e-05 | 0.00175 |
| Word/Phrase | Coefficient | Mythic Count | Non-mythic Count | p-value | q-value |
|---|---|---|---|---|---|
| οἱ | -1.8190 | 107 | 273 | 0.000118 | 0.00453 |
| ἄγαλμα | -1.6989 | 9 | 50 | 0.000223 | 0.00744 |
| τὰ | -1.4273 | 80 | 225 | 3.07e-05 | 0.00154 |
| σφισιν | -1.3006 | 16 | 43 | 0.116 | 0.382 |
| φωκέων | -1.2292 | 0 | 22 | 5.82e-05 | 0.00253 |
| μακεδόνων | -1.1589 | 0 | 25 | 2.07e-05 | 0.00115 |
| αὖθις | -1.0726 | 3 | 26 | 0.00284 | 0.0427 |
| πτολεμαῖον | -1.0550 | 0 | 18 | 0.000323 | 0.00872 |
| ἦν | -1.0456 | 26 | 76 | 0.0141 | 0.122 |
| μὴ | -1.0312 | 7 | 24 | 0.0901 | 0.338 |
| χαλκοῦν | -0.9905 | 1 | 23 | 0.000383 | 0.0101 |
| ἀλεξάνδρου | -0.9895 | 1 | 17 | 0.00523 | 0.0624 |
| ἀφροδίτης | -0.9368 | 1 | 13 | 0.0236 | 0.172 |
| μέν | -0.9174 | 6 | 30 | 0.00779 | 0.082 |
| μέν ἐστιν | -0.9116 | 0 | 13 | 0.00313 | 0.0454 |
| ναός | -0.9106 | 2 | 12 | 0.0985 | 0.354 |
| μάλιστα | -0.9079 | 16 | 55 | 0.00938 | 0.0928 |
| ἔργα | -0.9060 | 2 | 14 | 0.0647 | 0.269 |
| κεῖται | -0.9052 | 4 | 19 | 0.0791 | 0.314 |
| φιλίππου | -0.8995 | 0 | 12 | 0.00537 | 0.0624 |
| πτολεμαῖος | -0.8952 | 0 | 14 | 0.00329 | 0.0464 |
| τὸ ἔργον | -0.8765 | 1 | 17 | 0.00523 | 0.0624 |
| ἦσαν | -0.8566 | 4 | 24 | 0.0161 | 0.132 |
| ἀντιγόνου | -0.8538 | 0 | 16 | 0.000983 | 0.0209 |
| οὕτω | -0.8509 | 12 | 26 | 0.511 | 0.807 |
| ἀνάθημα | -0.8416 | 2 | 12 | 0.0985 | 0.354 |
| ἀνέθεσαν | -0.8384 | 3 | 22 | 0.0107 | 0.101 |
| ναυσὶν | -0.8197 | 5 | 24 | 0.0209 | 0.156 |
| ἀντὶ | -0.8096 | 5 | 13 | 0.425 | 0.731 |
| λίθου | -0.8087 | 2 | 16 | 0.0257 | 0.185 |