1
$\begingroup$

Are True Negatives always zero for a Named Entity Recognition task because TN in NER would mean a not entity being classified as Not entity?

Actual Entity [Microsoft Corp.] CEO [Steve Ballmer] announced the release of [Windows7] today

Model prediction [Microsoft Corp.] [CEO] [Steve] Ballmer announced the release of Windows7 [today]

What will be the TNs in the above example?

$\endgroup$
1
$\begingroup$

By definition, true negative is non-entity tokens in the text, agreed by gold standard. So;

TN: 5 = {announced, the, release, of}

But, in practice the number of true negative tokens are quite high and this heavily dominates the metrics, and any measure using TN (like accuracy) becomes highly controversial. Instead, preferred better metrics:

  • Recall: TP/(TP+FN) - coverage in all correct entities, and
  • Precision: TP/(TP+FP) - how reliable for a positive result
  • F1 Score (harmonic mean of precision and recall) can be used as a balanced metrics.
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.