Analyses › Lemma, excluding books 4 and 8, without rhetoric markers
Lemma, excluding books 4 and 8, without rhetoric markers
This model drops sentences tagged other and fits a balanced TF-IDF logistic regression to the remaining mythic and historical tags.
Model Performance Metrics
The following metrics are from the logistic regression classifier's performance on the test set:
| Class |
Precision |
Recall |
F1-Score |
Support |
| Historical |
0.636 |
0.621 |
0.629 |
248 |
| Mythic |
0.737 |
0.749 |
0.743 |
351 |
| Overall Accuracy |
0.696 |
599 |
Confusion Matrix
|
Predicted |
| Historical |
Mythic |
| Actual |
Historical |
154 |
94 |
| Mythic |
88 |
263 |
Counts
2,393mythic/historical sentences
1,404mythic
989historical
1,000features
Predictors of Mythic Sentences
Sort by:
| Word/Phrase |
English |
Coefficient |
Mythic Count |
Historical Count |
p-value |
q-value |
| θυγάτηρ |
|
1.8827 |
83 |
12 |
4.28e-10 |
2.14e-07 |
| δέ εἰμί |
|
1.7435 |
116 |
28 |
8.6e-09 |
2.15e-06 |
| εἰμί |
|
1.7318 |
596 |
284 |
4.86e-12 |
4.86e-09 |
| ἔποσ |
|
1.5755 |
28 |
0 |
4.5e-07 |
6.44e-05 |
| εἰμί ἐν |
|
1.4415 |
34 |
4 |
4.11e-05 |
0.00206 |
| τίκτω |
|
1.4324 |
21 |
3 |
0.00311 |
0.0536 |
| ἱδρύω |
|
1.3681 |
16 |
1 |
0.0021 |
0.0404 |
| αὐτόσ |
|
1.2664 |
382 |
226 |
0.0155 |
0.132 |
| ποιέω |
|
1.2002 |
156 |
91 |
0.128 |
0.429 |
| ὅτι |
|
1.1328 |
52 |
16 |
0.00176 |
0.0404 |
| ἱδρύομαι |
|
1.1015 |
10 |
0 |
0.00694 |
0.0812 |
| ἁρπάζω |
|
1.0978 |
19 |
1 |
0.00041 |
0.0138 |
| θάπτω |
|
1.0366 |
34 |
17 |
0.236 |
0.541 |
| εἰσ αὐτόσ |
|
1.0114 |
33 |
18 |
0.372 |
0.696 |
| ὁμολογέω |
|
0.9926 |
14 |
1 |
0.00642 |
0.0812 |
| ἔνθα |
|
0.9600 |
32 |
8 |
0.00384 |
0.0619 |
| χωρίον |
|
0.9578 |
36 |
10 |
0.00458 |
0.0692 |
| θρόνοσ |
|
0.9573 |
10 |
0 |
0.00694 |
0.0812 |
| καταμένω |
|
0.9403 |
10 |
1 |
0.0321 |
0.193 |
| συνοικέω |
|
0.9283 |
18 |
4 |
0.0292 |
0.18 |
| εἰμί γάρ |
|
0.8958 |
10 |
4 |
0.42 |
0.703 |
| καί ποιέω |
|
0.8921 |
16 |
4 |
0.0667 |
0.299 |
| ἐπονομάζω |
|
0.8799 |
9 |
0 |
0.0129 |
0.112 |
| σῶμα |
|
0.8787 |
6 |
0 |
0.0455 |
0.235 |
| ἀνάκειμαι |
|
0.8732 |
7 |
3 |
0.538 |
0.796 |
| δείκνυμι |
|
0.8727 |
11 |
3 |
0.175 |
0.461 |
| κομίζω |
|
0.8594 |
36 |
15 |
0.0748 |
0.318 |
| καί ὅτι |
|
0.8514 |
13 |
0 |
0.00115 |
0.0294 |
| γυνή |
|
0.8310 |
68 |
29 |
0.0175 |
0.141 |
| νομίζω |
|
0.8303 |
42 |
18 |
0.0661 |
0.299 |
Predictors of Historical Sentences
Sort by:
| Word/Phrase |
English |
Coefficient |
Mythic Count |
Historical Count |
p-value |
q-value |
| ὅσοσ |
|
-2.0262 |
23 |
49 |
3.41e-06 |
0.00031 |
| τυραννέω |
|
-1.6428 |
0 |
18 |
1.13e-07 |
1.88e-05 |
| νικάω |
|
-1.5956 |
8 |
24 |
9.89e-05 |
0.00412 |
| στρατιά |
|
-1.4991 |
6 |
27 |
1.51e-06 |
0.000168 |
| παρέχω |
|
-1.4475 |
3 |
15 |
0.00038 |
0.0136 |
| ἀφίστημι |
|
-1.4413 |
0 |
18 |
1.13e-07 |
1.88e-05 |
| πολύσ |
|
-1.4130 |
36 |
80 |
7.29e-10 |
2.43e-07 |
| πολεμέω |
|
-1.3775 |
8 |
28 |
7.06e-06 |
0.000588 |
| χρῆμα |
|
-1.3448 |
1 |
19 |
5.65e-07 |
7.06e-05 |
| ἐπί ἐγώ |
|
-1.2960 |
7 |
14 |
0.0188 |
0.148 |
| πλέον |
|
-1.2452 |
5 |
16 |
0.0011 |
0.029 |
| ἤδη |
|
-1.2402 |
23 |
31 |
0.0163 |
0.133 |
| πυνθάνομαι |
|
-1.2376 |
3 |
12 |
0.00295 |
0.0526 |
| καί εἰσ |
|
-1.2135 |
51 |
67 |
0.000535 |
0.0162 |
| τυραννίσ |
|
-1.2007 |
0 |
12 |
2.39e-05 |
0.00149 |
| ἀμύνω |
|
-1.1723 |
3 |
15 |
0.00038 |
0.0136 |
| εἶμι |
|
-1.1182 |
5 |
8 |
0.142 |
0.433 |
| πράσσω |
|
-1.1169 |
2 |
16 |
3.9e-05 |
0.00206 |
| ἐναντίοσ |
|
-1.0976 |
1 |
15 |
1.62e-05 |
0.00108 |
| καθαιρέω |
|
-1.0969 |
0 |
13 |
9.8e-06 |
0.000754 |
| μάλιστα |
|
-1.0839 |
29 |
51 |
4.04e-05 |
0.00206 |
| πείθω |
|
-1.0800 |
6 |
15 |
0.00508 |
0.0697 |
| ἐλπίζω |
|
-1.0778 |
1 |
13 |
8.49e-05 |
0.00369 |
| διαβαίνω |
|
-1.0777 |
8 |
21 |
0.000651 |
0.0192 |
| τύραννοσ |
|
-1.0775 |
0 |
6 |
0.00494 |
0.0692 |
| ἐμβάλλω |
|
-1.0717 |
4 |
17 |
0.000442 |
0.0138 |
| οἰκοδομέω |
|
-1.0598 |
9 |
19 |
0.00441 |
0.0689 |
| καί αὐτόσ |
|
-1.0514 |
40 |
51 |
0.00399 |
0.0634 |
| ὕστερον |
|
-1.0392 |
55 |
81 |
1.07e-05 |
0.000761 |
| στρατηγόσ |
|
-1.0330 |
2 |
16 |
3.9e-05 |
0.00206 |