Pausanias Analysis

Book-number regression from stylometric features

Can Stylometric Features Predict Book Number?

This page uses Ridge regression, not classification, to predict the numeric book number of each grammar-parsed sentence. Chance is the per-fold training-set mean book number.

These models use only sentences with current LLM grammar-token rows, so the sample is smaller than the full annotation corpus.
Strict three-way and mythic-vs-historical tasks drop rare both/mixed rows; one-vs-rest tasks use the available mythic/historical flags where present.
Character 4-grams are included as a traditional baseline; morphosyntax is the primary non-lexical stylometric family.

Open feature trend diagnostics | Download regression metrics CSV | Download regression features CSV

Regression Metrics

Variant	Features	n	Book counts	Features	Alpha	R2	MAE	Chance MAE	MAE improvement	RMSE	Status
Including Books 4 and 8	Morphosyntax	7,485	1: 877, 2: 823, 3: 644, 4: 733, 5: 652, 6: 570, 7: 618, 8: 974, 9: 745, 10: 849	240	3162.278	0.048	2.568	2.664	0.096	2.932	ready
Including Books 4 and 8	Function Words	7,485	1: 877, 2: 823, 3: 644, 4: 733, 5: 652, 6: 570, 7: 618, 8: 974, 9: 745, 10: 849	160	3162.278	0.034	2.594	2.664	0.070	2.954	ready
Including Books 4 and 8	Character 4-Grams	7,485	1: 877, 2: 823, 3: 644, 4: 733, 5: 652, 6: 570, 7: 618, 8: 974, 9: 745, 10: 849	220	3162.278	0.045	2.574	2.664	0.089	2.937	ready
Excluding Books 4 and 8	Morphosyntax	5,778	1: 877, 2: 823, 3: 644, 5: 652, 6: 570, 7: 618, 9: 745, 10: 849	240	3162.278	0.059	2.713	2.832	0.119	3.117	ready
Excluding Books 4 and 8	Function Words	5,778	1: 877, 2: 823, 3: 644, 5: 652, 6: 570, 7: 618, 9: 745, 10: 849	160	3162.278	0.040	2.749	2.832	0.084	3.149	ready
Excluding Books 4 and 8	Character 4-Grams	5,778	1: 877, 2: 823, 3: 644, 5: 652, 6: 570, 7: 618, 9: 745, 10: 849	220	3162.278	0.055	2.729	2.832	0.103	3.124	ready

Predictive Stylometric Features

Variant	Feature family	Largest standardized coefficients
Including Books 4 and 8	Morphosyntax	later_books: `Head>child PROPN>DET` (0.116) earlier_books: `Feature VerbForm=Part` (-0.098) earlier_books: `Head direction advcl:left` (-0.094) earlier_books: `Feature Voice=Act` (-0.082) later_books: `DepRel/UPOS nmod:PRON` (0.071) earlier_books: `DepRel/UPOS conj:NOUN` (-0.071) earlier_books: `Feature Gender=Masc` (-0.068) later_books: `UPOS PART` (0.066)
Including Books 4 and 8	Function Words	later_books: `Function form/UPOS οἱ:DET` (0.108) earlier_books: `Function form/UPOS οἱ:PRON` (-0.096) earlier_books: `Function DepRel advcl` (-0.086) later_books: `Function form/UPOS τε:CCONJ` (0.085) earlier_books: `Function DepRel obl` (-0.080) earlier_books: `Function UPOS PRON` (-0.079) later_books: `Function word ἡ` (0.076) later_books: `Function form/UPOS ἡ:DET` (0.073)
Including Books 4 and 8	Character 4-Grams	later_books: `char 4-gram ντο` (0.105) later_books: `char 4-gram δὴ` (0.101) earlier_books: `char 4-gram αὶ τ` (-0.100) later_books: `char 4-gram αὶ ἐ` (0.093) earlier_books: `char 4-gram ν ἀρ` (-0.093) later_books: `char 4-gram ἐκ` (0.092) later_books: `char 4-gram οἱ` (0.088) later_books: `char 4-gram ν τε` (0.080)
Excluding Books 4 and 8	Morphosyntax	later_books: `Head>child PROPN>DET` (0.115) later_books: `Feature Number=Plur` (0.111) earlier_books: `Head direction advcl:left` (-0.108) earlier_books: `Feature Number=Sing` (-0.106) earlier_books: `Feature VerbForm=Part` (-0.086) later_books: `Feature Voice=Mid` (0.085) later_books: `UPOS DET` (0.076) earlier_books: `Feature Voice=Act` (-0.075)
Excluding Books 4 and 8	Function Words	later_books: `Function form/UPOS οἱ:DET` (0.143) later_books: `Function form/UPOS τε:CCONJ` (0.110) earlier_books: `Function form/UPOS οἱ:PRON` (-0.108) earlier_books: `Function DepRel obl` (-0.098) earlier_books: `Function DepRel advcl` (-0.091) later_books: `Function word τε` (0.084) later_books: `Function word οἱ` (0.077) later_books: `Function word ἡ` (0.073)
Excluding Books 4 and 8	Character 4-Grams	later_books: `char 4-gram ντο` (0.171) earlier_books: `char 4-gram ν ἀρ` (-0.121) later_books: `char 4-gram δὴ` (0.111) later_books: `char 4-gram τε` (0.111) earlier_books: `char 4-gram ντοσ` (-0.110) later_books: `char 4-gram ἐκ` (0.102) later_books: `char 4-gram οἱ` (0.101) later_books: `char 4-gram τὰ` (0.100)