The 47-tag POS table

The POS system used by Bareun follows the standard of the National Institute of Korean Language. The number in the rightmost column of the table below ranges from 1 to 47. Internally, protocol buffers store the value as this number. Depending on how you output the result or which programming interface you use, it appears as a string.

By default, you can have it displayed as a string.

National Institute of Korean Language tagging guidelines

The morpheme tagging of Bareun follows the tagging guidelines of the National Institute of Korean Language and the Telecommunications Technology Association (TTA) "Standard Morpheme Tag Set (TTAK.KO-11.0010/R1)".

• National Institute of Korean Language morpheme tagging guidelines (PDF, Korean)
• TTA standard morpheme tag set (PDF, Korean)

Bareun's own tagging guidelines

Differences between the National Institute guidelines and the Bareun guidelines

For some word forms, Bareun set its own tagging guidelines and revised the training data. Where analysis was impossible, the data was revised in a reasonable way to make it analyzable. Bareun also accepts exceptions for variants of expressions that are common in spoken Korean, and was trained on data that was added (or revised) so those can be analyzed. Examples follow. (Urimalsaem is the NIKL Open Korean Dictionary.)

Spoken form dictionary