Zelma
One spelling. All preserved contexts.
Read claims alongside their sources.
- Script
- Latin
- Compound status
- Unknown
- In this record
- Relationships: 3 · observations: 1
Candidate data 1.0.0. Unknown values have not been filled with guesses.
Language information
Dictionary information and other contexts are kept separate. A statistical territory does not establish the language of a name; a classification does not determine a person's gender.
Dictionary contexts
None provided for this spelling in this version.
Other contexts
| Language | Gender in source | Basis | Evidence |
|---|---|---|---|
| Latvian | Not specified | Usage evidence | EvidenceLanguage: Latvian. Source category: Not specified. Basis: Usage evidence. This is a source claim, not a statement about a particular person. Open source and technical details →Technical details and JSONOriginal claim{
"basis": "usage_evidence",
"context_id": "ctx:3b88ea3c-80bb-5a45-b6b8-dfd51b434c65",
"country": null,
"evidence_id": "ev:0f1aa3a4-0f06-5f09-925c-ad1787c94ad4",
"form_id": "frm:1aa0dc4e-d763-5e49-9051-893d1aa94d06",
"gender": "unknown",
"language": "lv"
} |
| Not specified | Female | Source classification | EvidenceLanguage: Not specified. Source category: Female. Basis: Source classification. This is a source claim, not a statement about a particular person. Open source and technical details →Technical details and JSONOriginal claim{
"basis": "source_classification",
"context_id": "ctx:7fc44972-a38a-5fba-aa65-12a767436a54",
"country": null,
"evidence_id": "ev:a6d05c88-219a-5cca-86a1-6e8d4b588e2a",
"form_id": "frm:1aa0dc4e-d763-5e49-9051-893d1aa94d06",
"gender": "female",
"language": null
} |
| Hungarian | Not specified | Usage evidence | EvidenceLanguage: Hungarian. Source category: Not specified. Basis: Usage evidence. This is a source claim, not a statement about a particular person. Open source and technical details →Technical details and JSONOriginal claim{
"basis": "usage_evidence",
"context_id": "ctx:81829891-f408-53f5-8f6d-af6770a06308",
"country": null,
"evidence_id": "ev:c733c48a-4fda-5270-a178-a0c2a0c4e428",
"form_id": "frm:1aa0dc4e-d763-5e49-9051-893d1aa94d06",
"gender": "unknown",
"language": "hu"
} |
Related spellings
Every relationship has a direction and type. Source-reported equivalence does not imply universal interchangeability.
Relationships: 3
| From | To | Type | Evidence |
|---|---|---|---|
| Zelma | Selma | Source-reported equivalence | EvidenceThe source reports a relationship of type «Source-reported equivalence». The direction is shown in the table row. Open source and technical details →Technical details and JSONOriginal claim{
"evidence_id": "ev:14578899-a5b8-5cda-8505-ed7133e9120f",
"from_form_id": "frm:1aa0dc4e-d763-5e49-9051-893d1aa94d06",
"language": null,
"link_id": "lnk:4a1ee355-d9cd-5317-8536-41e597edb7b5",
"position": null,
"relation": "source_equivalence",
"scheme": null,
"to_form_id": "frm:c8167ff4-278d-5ace-9916-0caa1c43d6b8"
} |
| Zelma | Salome | Source-reported equivalence | EvidenceThe source reports a relationship of type «Source-reported equivalence». The direction is shown in the table row. Open source and technical details →Technical details and JSONOriginal claim{
"evidence_id": "ev:84c0b809-a926-59cd-a725-e8e945c75523",
"from_form_id": "frm:1aa0dc4e-d763-5e49-9051-893d1aa94d06",
"language": null,
"link_id": "lnk:781fa000-fc40-5aea-8111-a40cc4685521",
"position": null,
"relation": "source_equivalence",
"scheme": null,
"to_form_id": "frm:83436cd0-4bec-5535-81b9-67bd04de0d1b"
} |
| Zelma | Salmine | Source-reported equivalence | EvidenceThe source reports a relationship of type «Source-reported equivalence». The direction is shown in the table row. Open source and technical details →Technical details and JSONOriginal claim{
"evidence_id": "ev:78f264c6-3046-521f-badf-fb770a6dbae6",
"from_form_id": "frm:1aa0dc4e-d763-5e49-9051-893d1aa94d06",
"language": null,
"link_id": "lnk:72e02ffb-54ee-5099-bf4d-6467fd639d34",
"position": null,
"relation": "source_equivalence",
"scheme": null,
"to_form_id": "frm:8cb6a048-efbe-5031-a5c1-ff44fb47c6de"
} |
Usage in statistics
Series differ in coverage and counting units. Values cannot be added together to measure global popularity.
| Territory | Year | Category | Name position | Count |
|---|---|---|---|---|
| United States | 2025 | Female | Not available | 16 |
An unspecified position is not treated as first position. Full counting units, methods and qualifications are preserved in the original data.
Observation technical details
[
{
"country": "US",
"global_name_gender_inferred": false,
"method": "deterministic_sum_of_source_counts",
"published_count": 16,
"reported_sex": "female",
"source_csv_rows": [
2156934
],
"source_ref": "US",
"year": 2025
}
]A source behind every claim
Preserved evidence referenced by this record. Expand a description to view its source; original values are available in the technical details.
Usage context: Latvian · Not specified · Usage evidence
Usage context: Latvian · Not specified · Usage evidence
Open source and technical details →Usage context: Not specified · Female · Source classification
Usage context: Not specified · Female · Source classification
Open source and technical details →Usage context: Hungarian · Not specified · Usage evidence
Usage context: Hungarian · Not specified · Usage evidence
Open source and technical details →Provenance of the record «Zelma»
Provenance of the record «Zelma»
Open source and technical details →Relationship «Zelma → Selma»: Source-reported equivalence
Relationship «Zelma → Selma»: Source-reported equivalence
Open source and technical details →Relationship «Zelma → Salome»: Source-reported equivalence
Relationship «Zelma → Salome»: Source-reported equivalence
Open source and technical details →Relationship «Zelma → Salmine»: Source-reported equivalence
Relationship «Zelma → Salmine»: Source-reported equivalence
Open source and technical details →Full record
All fields, including qualifiers, language qualifications, original codes and unknown values.
Open the full record JSON →Field and compound-status provenance
{
"field_evidence": [
{
"characters": " -‐‑–'’",
"evidence_id": "ev:c2c552eb-2c77-52f8-82fe-a62876e4c4cb",
"field": "contains_name_separator",
"form_id": "frm:1aa0dc4e-d763-5e49-9051-893d1aa94d06",
"method": "deterministic",
"not_compound_classification": true,
"rule": "character-presence-001",
"value": false
}
],
"compound_claims": []
}Need the full dataset?
SQLite, JSONL and documentation for local use.
About the full package →