Text extraction
Also: parser, parsing
Pulling the words out of a file like a PDF so a program can use them.
To a computer this is a photo. It cannot search it or copy a sentence out of it.
The photo is read into real text. Now it can be searched, copied, and given to a model.
OCR makes mistakes on poor scans, handwriting, and tables. A missed page or a misread number travels silently into everything built on top of it.
The fuller definition
Text extraction obtains text from a digital document. Parsing interprets its structure, such as headings, tables, or fields, into a usable representation.
Go deeper
Delivery may be possible in April.
Subject to inspection.
Delivery may be possible in April. Subject to inspection.
source_id: A01
page: 2
version: original
This fictional example keeps a condition beside the proposition it qualifies. Extraction and chunking should preserve enough context to inspect what the source actually says.
A concrete example
A PDF parser extracts text and page references from a digitally created contract.
The distinction that matters
A PDF can contain selectable text yet yield a poor reading order or broken tables.
Related terms
OCR — Optical character recognition
Turning a picture of text, such as a scanned page, into real text a computer can search.
Documents & retrievalChunking
Cutting a long document into smaller pieces so they can be searched and fed to the model.
Documents & retrievalMetadata
Information about a file rather than its contents, such as its author, date, and page count.
Where it is taught
How document-grounded AI answers workSource and further reading
Sentence Transformers · Semantic searchThe definition and example are written for this library. The link gives background, official documentation or legal authority as labelled. Legal scope depends on the jurisdiction and current text; product features depend on release and settings.