Text Extraction

CURIE: gmeow:TextExtraction
IRI: https://blackcatinformatics.ca/gmeow/TextExtraction
Category: class
Defined by: gmeow:slices/provenance
Box roles: CBox role, TBox role (What is this?)

The text content extracted from a source object (e.g. a PDF attachment), linked to its source by gmeow:wasDerivedFrom. A faithful (ideally verbatim) rendering of the source's textual content, distinct from a gmeow:Summary, which condenses and is therefore lossy.

Structure

Subclass of: gmeow:Document

Practical Pattern

Use gmeow:TextExtraction as a specialized kind of gmeow:Document. Add statement metadata or a standpoint when the assertion needs provenance, confidence, or vantage.

Common Companion Terms

gmeow:Document

Usage Advice

Use when

Use for the text content pulled verbatim out of a source object — OCR or extraction from a PDF, image, or binary attachment — whose lineage to that source must be recorded.

Avoid when

Avoid for a condensed or paraphrased account (use gmeow:Summary) and for the source document itself; a TextExtraction is the derived text view, not the original artifact.

How to use

Type the extracted text as gmeow:TextExtraction, link it to its source with gmeow:wasDerivedFrom, and name the extracting activity via gmeow:wasGeneratedBy; record extraction confidence (OCR quality) on the statement layer.

Examples

ex:pdfText a gmeow:TextExtraction; gmeow:wasDerivedFrom ex:pdfAttachment.