Antonym Tagged Outputs

Antonym candidate files are generated annually and sent to linguists for tagging. This section show sampled tagged and processed outputs (in red) from the table in candidates section.

I. Tagged Outputs

II. Processed Outputs

• Antonyms

  The tagged file is processed by computer programs to generate antonyms if aPairs are canonical. The output file of antonyms has 9 fields. Each line represents an antonym pair (aPair) and includes all associated information. The table below shows the processed antonym outputs from above tagged file. aPair candidate [cashable|cashless] is removed because it is not an aPair in the Lexicon (not canonical).

  Ant-1	EUI-1	Ant-2	EUI-2	POS	Type	Negation	Domain	Source
  with	E0065516	without	E0065526	prep	B	N2	existence	LEX
  always	E0008403	never	E0042565	adv	UB	N2	frequency	LEX
  available	E0011401	unavailable	E0062924	adj	B	BN2	possibility	PD
  bacterial	E0011688	antibacterial	E0009417	adj	UB	O	quality	PD
  careful	E0015340	careless	E0015344	adj	UB	O	quality	SD
  cashable	E0794454	cashless	E0546479	adj	NA	O	DOMAIN_NONE	SD
  asleep	E0010731	awake	E0011449	adj	B	O	physical_property	CC
  junior	E0036311	senior	E0055206	noun	UB	O	temporal	CC
  heaven	E0030994	hell	E0031052	noun	UB	O	location	SN
  kid	E0036500	parent	E0045702	noun	UB	O	role	SN

• Negation Cue Words

  The tagged file is processed by computer programs to generate negation cue words if the negation field is tagged as [N1] or [N2]. The output file of antonyms has 3 fields. Each line represents a negation cue word, EUI and POS . Table below shows the processed negation cue word outputs from above tagged file.

  Negaton Cue Word	EUI	POS
  without	E0065526	prep
  never	E0042565	adv

Please refer to original design documents