Words and Phrases that Predict Mythic vs. Historical Content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 | 0.802 | 0.774 | 0.788 | 319 |
| Mythic | 0.779 | 0.806 | 0.793 | 315 |
| Overall Accuracy | 0.790 | 634 | ||
These words and phrases are most strongly associated with mythic content in Pausanias.
| Word/Phrase | Coefficient | Mythic Count | Non-mythic Count | p-value | q-value |
|---|---|---|---|---|---|
| λέγουσιν | 3.2232 | 256 | 61 | 1.66e-34 | 1.66e-31 |
| ἡρακλέους | 2.9946 | 106 | 11 | 1.31e-22 | 3.29e-20 |
| φασιν | 2.7536 | 191 | 40 | 3.98e-28 | 1.33e-25 |
| εἶναι | 2.6194 | 458 | 217 | 1.87e-28 | 9.33e-26 |
| ὅμηρος | 2.3326 | 77 | 7 | 1.74e-17 | 2.38e-15 |
| λέγουσι | 2.1672 | 148 | 41 | 1.9e-17 | 2.38e-15 |
| δʼ | 2.1648 | 145 | 66 | 2.57e-09 | 6.54e-08 |
| παῖδα | 1.8880 | 156 | 48 | 1.42e-16 | 1.42e-14 |
| ἡρακλέα | 1.8444 | 57 | 4 | 8.64e-14 | 4.8e-12 |
| ἡρακλεῖ | 1.8399 | 45 | 4 | 2.06e-10 | 6.64e-09 |
| ἡρακλῆς | 1.8187 | 62 | 6 | 4.98e-14 | 2.93e-12 |
| γενέσθαι | 1.7970 | 214 | 97 | 1.47e-13 | 7.76e-12 |
| θυγατέρα | 1.7658 | 82 | 9 | 2.33e-17 | 2.59e-15 |
| φασὶν | 1.7372 | 125 | 31 | 2.49e-16 | 2.07e-14 |
| ἀφικέσθαι | 1.7325 | 49 | 7 | 9e-10 | 2.57e-08 |
| ἀχιλλέως | 1.6694 | 34 | 1 | 5.32e-10 | 1.57e-08 |
| ἡρώων | 1.6687 | 29 | 4 | 4.16e-06 | 5.19e-05 |
| ἀγαμέμνονος | 1.6457 | 33 | 0 | 5.43e-11 | 2.01e-09 |
| ἔνθα | 1.6339 | 102 | 53 | 1.77e-05 | 0.000188 |
| παίδων | 1.6174 | 87 | 35 | 3.93e-07 | 6.45e-06 |
| πέλοπος | 1.6051 | 28 | 0 | 2e-09 | 5.25e-08 |
| ἐνταῦθα | 1.5922 | 296 | 170 | 1.33e-11 | 5.79e-10 |
| ἔχει | 1.5858 | 118 | 59 | 1.26e-06 | 1.75e-05 |
| αὐτὴν | 1.5562 | 141 | 62 | 1.36e-09 | 3.67e-08 |
| παῖδες | 1.5036 | 67 | 24 | 1.35e-06 | 1.85e-05 |
| θησέως | 1.4793 | 40 | 0 | 3.44e-13 | 1.72e-11 |
| φασι | 1.4478 | 55 | 9 | 4.02e-10 | 1.22e-08 |
| τοῦ | 1.3993 | 1016 | 934 | 2.09e-05 | 0.000218 |
| ἀνθρώπων | 1.3811 | 71 | 38 | 0.000609 | 0.00395 |
| hom | 1.3578 | 40 | 4 | 4.88e-09 | 1.14e-07 |
These words and phrases are most strongly associated with historical content in Pausanias.
| Word/Phrase | Coefficient | Mythic Count | Non-mythic Count | p-value | q-value |
|---|---|---|---|---|---|
| φιλίππου | -2.3001 | 3 | 60 | 8.45e-15 | 5.63e-13 |
| λακεδαιμόνιοι | -2.1265 | 15 | 110 | 2.18e-19 | 3.64e-17 |
| ἀχαιῶν | -2.0848 | 16 | 98 | 8.32e-16 | 6.4e-14 |
| λακεδαιμονίων | -1.9609 | 28 | 144 | 1.97e-20 | 3.93e-18 |
| ῥωμαίων | -1.8195 | 8 | 80 | 2.14e-16 | 1.94e-14 |
| δύο | -1.5977 | 54 | 93 | 0.00197 | 0.0101 |
| μακεδόνων | -1.5455 | 7 | 54 | 2.38e-10 | 7.44e-09 |
| αὖθις | -1.4927 | 61 | 117 | 3.88e-05 | 0.000374 |
| νῖκαι | -1.4828 | 2 | 34 | 1.93e-08 | 4.01e-07 |
| ὀλυμπίᾳ | -1.4826 | 19 | 89 | 3.62e-12 | 1.72e-10 |
| ἐν ὀλυμπίᾳ | -1.4774 | 18 | 85 | 9.7e-12 | 4.41e-10 |
| μάλιστα | -1.4627 | 164 | 233 | 0.000852 | 0.00522 |
| ἀνδρῶν | -1.4267 | 10 | 57 | 2.62e-09 | 6.54e-08 |
| λακεδαιμονίοις | -1.4192 | 26 | 96 | 1.01e-10 | 3.49e-09 |
| ἀθηναίων | -1.3521 | 39 | 105 | 3.42e-08 | 6.84e-07 |
| ἑλλάδα | -1.3463 | 7 | 47 | 1.32e-08 | 2.87e-07 |
| ἀνδρὸς | -1.3344 | 33 | 66 | 0.00126 | 0.00722 |
| λακεδαιμονίους | -1.2849 | 14 | 75 | 2.03e-11 | 8.47e-10 |
| μὴ | -1.2692 | 64 | 117 | 0.000122 | 0.00101 |
| ἐναντία | -1.2690 | 4 | 49 | 6.85e-11 | 2.45e-09 |
| πλέον | -1.2566 | 31 | 73 | 4.79e-05 | 0.000447 |
| τὴν ἑλλάδα | -1.2547 | 6 | 43 | 3.14e-08 | 6.4e-07 |
| κατʼ | -1.2335 | 23 | 63 | 1.76e-05 | 0.000188 |
| ἐπίγραμμα | -1.2154 | 29 | 55 | 0.00623 | 0.0253 |
| πολέμῳ | -1.1967 | 10 | 47 | 5.16e-07 | 8.19e-06 |
| μὲν | -1.1880 | 933 | 1039 | 0.00912 | 0.0342 |
| φίλιππος | -1.1864 | 1 | 29 | 5.74e-08 | 1.1e-06 |
| βασιλεὺς | -1.1612 | 14 | 47 | 2.18e-05 | 0.000222 |
| ἀλεξάνδρου | -1.1352 | 5 | 29 | 2.12e-05 | 0.000219 |
| τὴν εἰκόνα | -1.1345 | 4 | 31 | 3.45e-06 | 4.54e-05 |