Dataset Viewer
Auto-converted to Parquet Duplicate
lexeme
string
stem
string
sense
string
surface
string
count
int32
share
float32
method
string
source_corpus
string
base_text
string
hbo:0001
1
amaꞌ
60
0.4839
gloss
WLC
aaz_C01
hbo:0001
1
amaꞌ
30
0.1705
eflomal
WLC
aaz_C01
hbo:0001
1
aam
24
0.1935
gloss
WLC
aaz_C01
hbo:0001
1
in amaf
21
0.1694
gloss
WLC
aaz_C01
hbo:0001
1
in aamf ee
19
0.1532
gloss
WLC
aaz_C01
hbo:0001
1
in amaf
14
0.0795
eflomal
WLC
aaz_C01
hbo:0001
1
in aamf ee
12
0.0682
eflomal
WLC
aaz_C01
hbo:0001
1
naiꞌ
12
0.0682
eflomal
WLC
aaz_C01
hbo:0001
1
aam
10
0.0568
eflomal
WLC
aaz_C01
hbo:0001
1
amaf
9
0.0511
eflomal
WLC
aaz_C01
hbo:0001
1
in
8
0.0455
eflomal
WLC
aaz_C01
hbo:0001
1
aamf ee
8
0.0455
eflomal
WLC
aaz_C01
hbo:0001
1
amaꞌ in
6
0.0341
eflomal
WLC
aaz_C01
hbo:0001
1
hit amaꞌ
6
0.0341
eflomal
WLC
aaz_C01
hbo:0001
1
ee
6
0.0341
eflomal
WLC
aaz_C01
hbo:0001
1
hai amaꞌ
6
0.0341
eflomal
WLC
aaz_C01
hbo:0001
1
mfain
4
0.0227
eflomal
WLC
aaz_C01
hbo:0001
1
minaꞌ
3
0.017
eflomal
WLC
aaz_C01
hbo:0001
1
amaꞌ naꞌbees neu
3
0.017
eflomal
WLC
aaz_C01
hbo:0001
1
uuf
2
0.0114
eflomal
WLC
aaz_C01
hbo:0001
1
unuꞌ
2
0.0114
eflomal
WLC
aaz_C01
hbo:0001
1
aamf ee in
2
0.0114
eflomal
WLC
aaz_C01
hbo:0001
1
hit
2
0.0114
eflomal
WLC
aaz_C01
hbo:0001
1
neu in amaf
2
0.0114
eflomal
WLC
aaz_C01
hbo:0001
1
au
2
0.0114
eflomal
WLC
aaz_C01
hbo:0001
1
ho amaꞌ
2
0.0114
eflomal
WLC
aaz_C01
hbo:0001
1
neu
2
0.0114
eflomal
WLC
aaz_C01
hbo:0001
1
hit amaꞌ in
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
aam honif
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
hai
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
amaf neu
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
in aamf ee te
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
noik aamf
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
te amaꞌ
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
amaꞌ te
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
teas
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
neu in aamf ee
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
aam uuf neu atoin
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
aamf
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
ee te hai amaꞌ
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
nmaet
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
nasaeb
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
in amaf naiꞌ
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
natoon
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
atukus
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
in aamf
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
in aamf ee nameen
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
muusn
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
naꞌbees
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
tataf
1
0.0057
eflomal
WLC
aaz_C01
hbo:0001
1
amaf
1
1
gapfill
WLC
aaz_C01
hbo:0014
qal
1
nroim
2
1
eflomal
WLC
aaz_C01
hbo:0014
qal
1
nroim
2
1
gloss
WLC
aaz_C01
hbo:0028
1
efer
1
1
eflomal
WLC
aaz_C01
hbo:0028
1
efer
1
1
gloss
WLC
aaz_C01
hbo:0039
1
abimaer
1
1
eflomal
WLC
aaz_C01
hbo:0039
1
abimaer
1
1
gloss
WLC
aaz_C01
hbo:0040
1
abimerek
9
0.5
gloss
WLC
aaz_C01
hbo:0040
1
uisf
8
0.4444
gloss
WLC
aaz_C01
hbo:0040
1
abimerek
5
0.2632
eflomal
WLC
aaz_C01
hbo:0040
1
uisf
4
0.2105
eflomal
WLC
aaz_C01
hbo:0040
1
usif abimerek
4
0.2105
eflomal
WLC
aaz_C01
hbo:0040
1
mneit neu usif
1
0.0526
eflomal
WLC
aaz_C01
hbo:0040
1
ngguin
1
0.0526
eflomal
WLC
aaz_C01
hbo:0040
1
usif
1
0.0526
eflomal
WLC
aaz_C01
hbo:0040
1
ahh
1
0.0526
eflomal
WLC
aaz_C01
hbo:0040
1
nakain
1
0.0526
eflomal
WLC
aaz_C01
hbo:0040
1
abimerek namnaub
1
0.0526
eflomal
WLC
aaz_C01
hbo:0040
1
ahh
1
0.0556
gloss
WLC
aaz_C01
hbo:0046
1
anpao
1
1
eflomal
WLC
aaz_C01
hbo:0046
1
anpao
1
1
gloss
WLC
aaz_C01
hbo:0056
hithpael
1
meꞌur
1
1
eflomal
WLC
aaz_C01
hbo:0056
hithpael
1
meꞌur
1
1
gloss
WLC
aaz_C01
hbo:0057
2
ꞌnikaꞌ
1
1
eflomal
WLC
aaz_C01
hbo:0057
2
ꞌnikaꞌ
1
1
gloss
WLC
aaz_C01
hbo:0060
1
nbeꞌen
2
0.5
gloss
WLC
aaz_C01
hbo:0060
1
ꞌpao beeꞌt
1
0.25
eflomal
WLC
aaz_C01
hbo:0060
1
fai
1
0.25
eflomal
WLC
aaz_C01
hbo:0060
1
nbeꞌen
1
0.25
eflomal
WLC
aaz_C01
hbo:0060
1
masiras
1
0.25
eflomal
WLC
aaz_C01
hbo:0060
1
ꞌpao beeꞌt
1
0.25
gloss
WLC
aaz_C01
hbo:0060
1
masiras
1
0.25
gloss
WLC
aaz_C01
hbo:0067
1
abel misraim
1
1
eflomal
WLC
aaz_C01
hbo:0067
1
abel misraim
1
1
gloss
WLC
aaz_C01
hbo:0068
1
fatu
6
0.6
gloss
WLC
aaz_C01
hbo:0068
1
fatu
4
0.2857
eflomal
WLC
aaz_C01
hbo:0068
1
faut kouꞌ goes
2
0.1429
eflomal
WLC
aaz_C01
hbo:0068
1
faut
2
0.2
gloss
WLC
aaz_C01
hbo:0068
1
faut arimat
1
0.0714
eflomal
WLC
aaz_C01
hbo:0068
1
kouꞌ goes
1
0.0714
eflomal
WLC
aaz_C01
hbo:0068
1
faut goes
1
0.0714
eflomal
WLC
aaz_C01
hbo:0068
1
faut akaꞌnunuꞌ
1
0.0714
eflomal
WLC
aaz_C01
hbo:0068
1
naꞌroob fatu
1
0.0714
eflomal
WLC
aaz_C01
hbo:0068
1
ntoeb
1
0.0714
eflomal
WLC
aaz_C01
hbo:0068
1
faut tobef
1
0.0714
eflomal
WLC
aaz_C01
hbo:0068
1
uꞌpeis
1
0.0714
eflomal
WLC
aaz_C01
hbo:0068
1
faut goes
1
0.1
gloss
WLC
aaz_C01
hbo:0068
1
faut akaꞌnunuꞌ
1
0.1
gloss
WLC
aaz_C01
hbo:0085
1
abraham
94
1
gloss
WLC
aaz_C01
hbo:0085
1
naiꞌ abraham
67
0.6569
eflomal
WLC
aaz_C01
End of preview. Expand in Data Studio

senses_attested — attested target renderings per lexeme sense

The empirical evidence layer produced for shoresh (bcv-query data-contract): for a lexeme in a disambiguated (binyan, sense), which target-language words attest it, with counts. It is the supply that fills shoresh's senses_i18n/_gaps demand and cross-checks the llm_strongs_glosses predictions — it does not replace shoresh's curated senses_i18n/<iso>.tsv; consumed as an HF Parquet dataset.

Schema (per row)

column meaning
lexeme MACULA lexeme (the anchor), e.g. hbo:0006
stem MACULA binyan (qal/piel/hiphil/…); empty for non-verbs
sense sense number (ordinal) — see licensing
surface attested target rendering (lowercased)
count times this (lexeme, stem, sense) → surface was aligned
share count / Σ count for that (lexeme, stem, sense) within one base_text
method alignment method (eflomal)
source_corpus the original Hebrew corpus (e.g. WLC)
base_text the target edition attested (e.g. ind_C01) — the per-row provenance dimension

Key: (lexeme, stem, sense) — MACULA lexeme (anchor; BHSA lex dropped) + MACULA binyan + sense number, read inline from the enriched lexeme-spine.db. OT/Hebrew only (senses are Hebrew; Greek tokens carry none).

Multi-version: base_text is per-row, so several translations of a language are pooled into one iso=<lang> partition — a union of per-edition runs, each row tagged by edition; share stays per-edition. Cross-edition agreement (how many base_texts attest a given (lexeme,stem,sense)→ surface) is the confidence signal, derivable directly from the rows. (Swedish iso=swe pools swe_fol Folkbibeln + swe_svk Kärnbibeln.)

Removal / takedown policy

Each row is a per-edition attestation carrying its base_text, so a rights-holder can request removal and it is a clean row-drop + republish (the dataset is content-addressed via each partition's content_sha256). Because most (lexeme,stem,sense)→surface facts are attested by more than one edition, dropping one edition typically leaves the linguistic fact intact via the others — properly attributed. Rows are never re-emitted with provenance stripped: a removed attestation is removed, not anonymized.

Removals are driven by a committed, auditable config: data/senses_exclude.json (read automatically on every build). A row is dropped if it matches any rule; a rule matches when all its stated fields equal the row's — fields lexeme, stem, sense, surface, base_text, omit to wildcard:

{"exclude": [
  {"base_text": "swe_fol"},                       // drop a whole edition
  {"base_text": "swe_fol", "surface": "herren"}   // drop one surface within an edition
]}

After exclusion, survivor shares renormalise (per edition), so a removed row leaves no residue; the manifest records excluded: {rules, rows_dropped} for the audit trail. To action a takedown: add a rule, re-run senses_attested for the affected language, republish.

Licensing — CC-BY-4.0, deliberately label-free

The key is MACULA-derived (lexeme + binyan), so this dataset is CC-BY-4.0 — attribute Clear-Bible MACULA. We carry the sense number only and no English sense label: shoresh's sense labels are UBS-MARBLE "used with permission" (not redistributable), so the payload is pure attestation (lexeme, stem, sense#, surface, count) — CC-BY clean. Regenerate: python -m lexeme_aligner.senses_attested --iso <iso> --method eflomal. Same git-ignored-Parquet + committed-manifest.json layout as lexeme-alignments.

Downloads last month
571