Artifex uses two label families: period (ordered date bins) and region (a hierarchical cultural-geographic taxonomy). Both are versioned through configuration files; changing them changes the dataset version.

Dates

Internal convention

• Years are integers on the astronomical scale: 1 CE = 1, 1 BCE = 0, 2 BCE = −1, n BCE = 1 − n. This is the only scale on which subtraction across the era boundary is correct (there is no historical year 0: from 50 BCE to 50 CE is 99 years).
• The Met and Cleveland APIs give BCE years as negative historical years (500 BCE = −500). They are converted with historical_to_astronomical (src/artifex/dates.py).
• Every date is a closed range [date_start, date_end]. A single year is a range of length zero.
• The museum's own date text is preserved in original_date_text; flags date_approximate ("c.", "about", "late …") and date_uncertain ("?", "probably", "or", …) are derived from it.
• Unusable values are kept as unknown with a reason in date_issue: missing, zero-placeholder (the Met's 0/0 for undated objects), out-of-range, or reversed-range (start and end swapped and fixed).

Display

User-facing dates show the museum's text where available. Normalised ranges are formatted with explicit eras where needed: "500–100 BCE", "50 BCE–50 CE", "800 CE", "1600–1650".

Period bins

Defined in configs/dataset.yaml (period.bins); contiguous, half-open intervals [start, end):

Bins are deliberately neutral date ranges, not style periods ("Renaissance", "Edo"), because the collection spans traditions whose periodisations do not align. Bins are narrower where the collection is dense and wider where it is sparse; a bin with fewer than min_records_per_class labelled works in the modelling subset is not a classifier target (its works remain in the collection and the retrieval index).

Assignment rule. A work is labelled with a bin only if at least 75 % of its documented range lies inside that bin (assign_period in src/artifex/labels.py). Otherwise period_status is spans-bins and the work has no period label. It is still evaluated with range-aware metrics. The rule avoids forcing "16th–17th century" into one bin.

Regions

Principle

A region label describes the cultural-geographic attribution recorded by the holding museum for the artwork (its culture or place of production). It is not the place depicted and, where avoidable, not the artist's birthplace. Regions are historical-geographic zones, not modern states; modern country names in museum records only point into these zones.

Hierarchy (configs/region_mapping.yaml)

Only broad regions (and traditions) with enough labelled works become classifier targets; the deployed list is shown on the Methodology page under Current configuration and with every analysis.

Mapping procedure

1. Fields are consulted in order: object_culture → object_geography → artist_nationality → department_hint (the last is used only for the Met's Egyptian Art department, which is itself a geographic attribution).
2. All rules are tested against the string. If rules from two different broad regions match, the record is ambiguous and stays unlabelled — the search does not fall back to a weaker field. Examples: "United States or England", "Western Tibet [or Kashmir (?)]".
3. override rules handle compound terms whose meaning is not the union of their words ("Roman Egypt" → Nile Valley; "British India" → South Asia; "Sino-Tibetan" → Himalayan region).
4. When a culture string names the place of the painting explicitly ("…Vikramashila Monastery. Painting: Nepal, Kathmandu"; "Herat (text); Mughal India (paintings)"), only that clause is mapped.
5. A sub-region is set only if exactly one tradition matches.
6. Strings matching nothing are unmapped. All distinct strings, their counts and mappings are written to data/manifests/region_mapping_report.csv for review.

Every record keeps culture_original, region_original, region_basis, region_status and region_matched_text.

Known tensions

• Artist nationality is used for many European paintings because the museums record no culture for them. It is a conventional "school" attribution, not a place of production (a Dutch artist working in Paris is labelled Low Countries).
• Dynastic or courtly labels (Mughal, Safavid) are mapped to the region of the court; works made for one court by artists from elsewhere follow the museum's attribution.
• Himalayan and Indian Buddhist painting overlap culturally; strings naming both are left ambiguous.
• Modern borders (Pakistan, Afghanistan) appear in museum records for historical regions; they are mapped into the broad zones above, which are intentionally coarse.