Hugging Face's logo

Datasets:
bcv-commons
/
argument-roles

Modalities:
Text
Formats:
parquet
Languages:
Ancient Hebrew
Official Aramaic (700-300 BCE)
Ancient Greek (to 1453)
Size:
100K - 1M
Tags:
bible
hebrew
greek
semantic-roles
rrg
prepositions
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
 Community
larsgson commited on
Commit
e7da4d5
·
verified ·
1 Parent(s): 6e8d9be

Argument roles v0.1: relation, voice, macrorole and semantic role per clause constituent (CC BY-SA 4.0)

Browse files
Files changed (6) hide show
  1. README.md +84 -0
  2. SHA256SUMS +5 -0
  3. build.json +56 -0
  4. checks.md +80 -0
  5. data/greek-nt.parquet +3 -0
  6. data/hebrew.parquet +3 -0
README.md ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-sa-4.0
3
+ language: [hbo, arc, grc]
4
+ pretty_name: Argument roles for the Hebrew and Greek clause layouts
5
+ tags: [bible, hebrew, greek, semantic-roles, rrg, prepositions, louw-nida, septuagint]
6
+ configs:
7
+ - config_name: hebrew
8
+ data_files: data/hebrew.parquet
9
+ - config_name: greek-nt
10
+ data_files: data/greek-nt.parquet
11
+ ---
12
+
13
+ # Argument roles (v0.1)
14
+
15
+ Every constituent of every clause of the two bcv-commons clause layouts, with its grammatical relation, the clause's voice, the verb
16
+ frame's macrorole and a semantic role: actor, undergoer, recipient, addressee, goal, source, location, time, accompaniment,
17
+ instrument, means, beneficiary, agent and others. These are what a translation into another language needs to choose case markers,
18
+ prepositions and voice.
19
+
20
+ | config | layout (pinned) | constituents | with a role | preposition phrases with a role |
21
+ |---|---|---|---|---|
22
+ | `hebrew` | [bcv-commons/hebrew-clause-layout](https://huggingface.co/datasets/bcv-commons/hebrew-clause-layout) v0.37.1 (`32f89b71`) | 146,733 | 75,896 | 28,597 of 55,735 |
23
+ | `greek-nt` | [bcv-commons/greek-nt-clause-layout](https://huggingface.co/datasets/bcv-commons/greek-nt-clause-layout) v0.7 (`9168fac3`) | 48,186 | 25,662 | 7,837 of 9,632 |
24
+
25
+ Join a row to its clause by `clause_key`: the Hebrew layout's `clause_key`, the Greek layout's `first`, as in
26
+ [bcv-commons/clause-columns](https://huggingface.co/datasets/bcv-commons/clause-columns).
27
+
28
+ ## Columns
29
+
30
+ | column | in | description |
31
+ |---|---|---|
32
+ | `clause_key` | both | the layout clause |
33
+ | `words` | both | the constituent's word ids inside the clause, in text order, space-joined |
34
+ | `head` | both | its first noun, pronoun, adjective or number word; empty when none |
35
+ | `relation` | both | the constituent's role in MACULA's tree: s, o, o2, io, p, adv, pp |
36
+ | `verb`, `voice` | both | the clause's main verb and voice (clause-columns) |
37
+ | `macrorole` | both | the verb frame's label for the constituent (MACULA frames: A0, A1, A2, AA); empty when the frame does not list it |
38
+ | `case` | greek-nt | the head's case |
39
+ | `prep`, `prep_lemma` | both | the preposition heading the constituent |
40
+ | `prep_ln` | greek-nt | the Louw-Nida subdomain of that preposition's sense here (bcv-commons/word-senses) |
41
+ | `prep_lxx` | hebrew | the Septuagint word MACULA aligns to the Hebrew preposition |
42
+ | `lxx_case` | hebrew | the case of the Septuagint word aligned to the phrase's first noun (GLAUx) |
43
+ | `role` | both | the semantic role; empty where the evidence does not decide |
44
+ | `role_basis` | both | which rule gave it (below) |
45
+
46
+ ## How the roles are assigned
47
+
48
+ - **frame / frame_passive:** the verb frame's A0 is the actor, A1 the undergoer (frame_passive: the passive subject), A2 kept as A2.
49
+ - **ln (Greek):** the Louw-Nida subdomain of the preposition's sense names the role. For example, 90.A agent, 90.B instrument,
50
+ 84.A source, 84.B goal, 83 location, 67 time, 89.H purpose, 34 accompaniment. With a verb of speaking (Louw-Nida 33), an
51
+ experiencer or goal is the addressee (`ln_speech`, `+speech`).
52
+ - **Hebrew preposition phrases,** in this order:
53
+ - `lxx_nt_domain`: the Septuagint's preposition, the case of its noun, and the Louw-Nida domain that noun's lemma has in the
54
+ New Testament. These are read through the New Testament's own (preposition, case, domain) → role table, with at least 8
55
+ examples and 60% agreement.
56
+ - `lxx_nt`: the same without the domain (at least 20 examples, 60%).
57
+ - `lxx_dative`: לְ that the Septuagint gives as a bare dative is the recipient.
58
+ - `hebrew_prep`: כְּ is a comparison (`standard`).
59
+ - `hebrew_domain`: the Hebrew preposition with its noun's SDBH domain, learned from the rows above.
60
+ - `+speech`: a goal with אֶל, לְ or עַל after a verb of speaking (אָמַר, דָּבַר, קָרָא, צָוָה, עָנָה ...) is the addressee.
61
+
62
+ Nothing comes from BHSA or TBTA, and no English text is used as evidence. No row was checked by hand; a human review of samples is
63
+ planned and will be reported here.
64
+
65
+ ## Measured use
66
+
67
+ A rule-based drafting engine was tested on the Tanakh in six rotating test groups, each drafted by an engine that had not seen it.
68
+ The figure is the share of right choices where the role changed the engine's choice:
69
+
70
+ | use | Indonesian | Swahili | Tagalog |
71
+ |---|---|---|---|
72
+ | prepositions chosen by Hebrew preposition and role | +7.4 | +16.2 | +5.4 |
73
+ | Tagalog case markers (ang / ng) by actor and undergoer | | | +3.6 |
74
+
75
+ Word-level F1 of the whole draft against the published translation rises by +0.46 (Indonesian) and +0.85 (Swahili) from the
76
+ prepositions alone. The Tagalog figure (+0.20) also includes kay / sa chosen by role and by whether the noun refers to a person.
77
+
78
+ The Greek side has not been measured in drafting yet.
79
+
80
+ ## Licence
81
+
82
+ CC BY-SA 4.0: the Hebrew layout's clause keys, the UBS senses and the GLAUx parses are CC BY-SA 4.0. Credit: MACULA (Clear Bible /
83
+ Biblica), United Bible Societies (word senses), GLAUx (KU Leuven, Alek Keersmaekers), make-tanakh-research (the measurements), and
84
+ this dataset.
SHA256SUMS ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ 2bf06bc54ab89541f25dc77df7a38f4918f93b942e4ea4677eb94921f56c4496 README.md
2
+ bef71283540cc59b489e427a494e070e317471e44f6dd657316bcd7e01c522b6 build.json
3
+ 181de5f325f20afc891f6c4b28d469b046541b18dc2095fd8d4a90a541643413 checks.md
4
+ 2c44ee569da6fa2f3082059e14d88b0405d0fb780f1605314b3b829b7157b366 data/greek-nt.parquet
5
+ 80d798a966f766320967f135ee309e877d4e9ec0328a671b30fd00c3bc44d6e0 data/hebrew.parquet
build.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema": 1,
3
+ "version": "v0.1",
4
+ "built": "2026-11-13",
5
+ "code": "fc06571",
6
+ "licence": "CC BY-SA 4.0",
7
+ "layouts": {
8
+ "hebrew": {
9
+ "repo": "bcv-commons/hebrew-clause-layout",
10
+ "version": "v0.37.1",
11
+ "revision": "32f89b719c0fc648320428c8492709c474c17ff2",
12
+ "licence": "CC BY-SA 4.0",
13
+ "key": "clause_key"
14
+ },
15
+ "greek-nt": {
16
+ "repo": "bcv-commons/greek-nt-clause-layout",
17
+ "version": "v0.7",
18
+ "revision": "9168fac3bd26d7d0e4e63e3a5905199c2182905a",
19
+ "licence": "CC BY 4.0",
20
+ "key": "first"
21
+ }
22
+ },
23
+ "inputs": {
24
+ "clause-columns": "bcv-commons/clause-columns v1.0 (verb, voice)",
25
+ "word-senses": "bcv-commons/word-senses v0.1",
26
+ "lxx": "GLAUx Septuagint parses (CC BY-SA 4.0), spine-releases lxx-glaux-2026-10-10",
27
+ "macula": "MACULA Hebrew and Greek trees, frames, Septuagint alignment column (CC BY 4.0)"
28
+ },
29
+ "not_read": [
30
+ "BHSA",
31
+ "TBTA",
32
+ "English text as evidence"
33
+ ],
34
+ "counts": {
35
+ "hebrew": {
36
+ "rows": 146733,
37
+ "clauses": 77605,
38
+ "with_role": 75896,
39
+ "prep": 55735,
40
+ "prep_with_role": 28597
41
+ },
42
+ "greek-nt": {
43
+ "rows": 48186,
44
+ "clauses": 26463,
45
+ "with_role": 25662,
46
+ "prep": 9632,
47
+ "prep_with_role": 7837
48
+ }
49
+ },
50
+ "files": {
51
+ "README.md": "2bf06bc54ab89541f25dc77df7a38f4918f93b942e4ea4677eb94921f56c4496",
52
+ "checks.md": "181de5f325f20afc891f6c4b28d469b046541b18dc2095fd8d4a90a541643413",
53
+ "data/greek-nt.parquet": "2c44ee569da6fa2f3082059e14d88b0405d0fb780f1605314b3b829b7157b366",
54
+ "data/hebrew.parquet": "80d798a966f766320967f135ee309e877d4e9ec0328a671b30fd00c3bc44d6e0"
55
+ }
56
+ }
checks.md ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Argument roles per constituent: checks
2
+
3
+ | corpus | count | value |
4
+ |---|---|---|
5
+ | grc | 48,186 | rows |
6
+ | grc | 22,524 | role:- |
7
+ | grc | 9,632 | prep |
8
+ | grc | 9,597 | role:undergoer |
9
+ | grc | 7,799 | prep_with_role |
10
+ | grc | 5,848 | role:actor |
11
+ | grc | 2,418 | role:A2 |
12
+ | grc | 1,795 | prep_no_role |
13
+ | grc | 1,199 | role:location |
14
+ | grc | 1,144 | role:goal |
15
+ | grc | 726 | role:time |
16
+ | grc | 548 | role:accompaniment |
17
+ | grc | 344 | role:agent |
18
+ | grc | 326 | role:content |
19
+ | grc | 290 | role:means |
20
+ | grc | 289 | role:source |
21
+ | grc | 282 | role:specification |
22
+ | grc | 268 | role:reason |
23
+ | grc | 254 | role:source_event |
24
+ | grc | 221 | role:state |
25
+ | grc | 216 | role:purpose |
26
+ | grc | 213 | role:experiencer |
27
+ | grc | 209 | role:separation |
28
+ | grc | 200 | role:addressee |
29
+ | grc | 166 | role:beneficiary |
30
+ | grc | 154 | role:standard |
31
+ | grc | 110 | role:opposition |
32
+ | grc | 81 | role:circumstance |
33
+ | grc | 75 | role:origin |
34
+ | grc | 71 | role:instrument |
35
+ | grc | 66 | role:result |
36
+ | grc | 63 | role:manner |
37
+ | grc | 59 | role:partitive |
38
+ | grc | 54 | role:reason_participant |
39
+ | grc | 51 | role:path |
40
+ | grc | 51 | role:viewpoint |
41
+ | grc | 38 | prep_core |
42
+ | grc | 32 | role:distribution |
43
+ | grc | 18 | role:guarantor |
44
+ | grc | 11 | role:basis |
45
+ | grc | 8 | role:responsibility |
46
+ | hbo | 146,733 | rows |
47
+ | hbo | 70,837 | role:- |
48
+ | hbo | 55,735 | prep |
49
+ | hbo | 28,322 | prep_with_role |
50
+ | hbo | 27,133 | prep_no_role |
51
+ | hbo | 26,029 | role:actor |
52
+ | hbo | 21,279 | role:undergoer |
53
+ | hbo | 12,934 | basis:hebrew_domain |
54
+ | hbo | 6,663 | role:location |
55
+ | hbo | 5,224 | role:recipient |
56
+ | hbo | 5,224 | basis:lxx_dative |
57
+ | hbo | 4,081 | role:standard |
58
+ | hbo | 3,488 | role:goal |
59
+ | hbo | 3,321 | role:time |
60
+ | hbo | 2,519 | basis:hebrew_prep |
61
+ | hbo | 1,248 | role:addressee |
62
+ | hbo | 1,110 | role:source |
63
+ | hbo | 982 | role:accompaniment |
64
+ | hbo | 800 | role:separation |
65
+ | hbo | 775 | role:source_event |
66
+ | hbo | 280 | prep_core |
67
+ | hbo | 266 | role:A2 |
68
+ | hbo | 162 | role:result |
69
+ | hbo | 121 | role:content |
70
+ | hbo | 85 | role:opposition |
71
+ | hbo | 60 | role:beneficiary |
72
+ | hbo | 49 | role:state |
73
+ | hbo | 47 | role:agent |
74
+ | hbo | 39 | role:specification |
75
+ | hbo | 16 | role:means |
76
+ | hbo | 15 | role:circumstance |
77
+ | hbo | 14 | role:reason_participant |
78
+ | hbo | 12 | role:reason |
79
+ | hbo | 8 | role:experiencer |
80
+ | hbo | 2 | role:purpose |
data/greek-nt.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2c44ee569da6fa2f3082059e14d88b0405d0fb780f1605314b3b829b7157b366
3
+ size 1408803
data/hebrew.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80d798a966f766320967f135ee309e877d4e9ec0328a671b30fd00c3bc44d6e0
3
+ size 4566778